Skip to content

Instantly share code, notes, and snippets.

@tej87681088

tej87681088/TEJAPI_Medium資料科學13.ipynb

Created April 20, 2022 02:59
Show Gist options
  • Save tej87681088/04373fd8987a38e9487757db2ee5f729 to your computer and use it in GitHub Desktop.
Save tej87681088/04373fd8987a38e9487757db2ee5f729 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
TEJAPI_Medium資料科學13.ipynb
{
"cells": [
{
"cell_type": "markdown",
"id": "8cb19eab",
"metadata": {},
"source": [
"## Import Packages"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "5c011793",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"\n",
"Bad key The font.family property has five values in file /Users/wujinru/opt/anaconda3/lib/python3.8/site-packages/matplotlib/mpl-data/matplotlibrc, line 211 ('The font.family property has five values:')\n",
"You probably need to get an updated matplotlibrc file from\n",
"https://github.com/matplotlib/matplotlib/blob/v3.3.4/matplotlibrc.template\n",
"or from the matplotlib source distribution\n",
"\n",
"Bad key The font.family property has five values in file /Users/wujinru/opt/anaconda3/lib/python3.8/site-packages/matplotlib/mpl-data/matplotlibrc, line 211 ('The font.family property has five values:')\n",
"You probably need to get an updated matplotlibrc file from\n",
"https://github.com/matplotlib/matplotlib/blob/v3.3.4/matplotlibrc.template\n",
"or from the matplotlib source distribution\n"
]
}
],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline\n",
"import seaborn as sns\n",
"sns.set()\n",
"\n",
"import tejapi\n",
"tejapi.ApiConfig.api_key = 'Your Key'\n",
"tejapi.ApiConfig.ignoretz = True"
]
},
{
"cell_type": "markdown",
"id": "5ebe06cf",
"metadata": {},
"source": [
"## Import Data"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "f5e24bf3",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['1101',\n",
" '1102',\n",
" '1103',\n",
" '1104',\n",
" '1108',\n",
" '1109',\n",
" '1110',\n",
" '1201',\n",
" '1203',\n",
" '1210',\n",
" '1213',\n",
" '1215',\n",
" '1216',\n",
" '1217',\n",
" '1218',\n",
" '1219',\n",
" '1220',\n",
" '1225',\n",
" '1227',\n",
" '1229',\n",
" '1231',\n",
" '1232',\n",
" '1233',\n",
" '1234',\n",
" '1235',\n",
" '1236',\n",
" '1256',\n",
" '1301',\n",
" '1303',\n",
" '1304',\n",
" '1305',\n",
" '1307',\n",
" '1308',\n",
" '1309',\n",
" '1310',\n",
" '1312',\n",
" '1313',\n",
" '1314',\n",
" '1315',\n",
" '1316',\n",
" '1319',\n",
" '1321',\n",
" '1323',\n",
" '1324',\n",
" '1325',\n",
" '1326',\n",
" '1337',\n",
" '1338',\n",
" '1339',\n",
" '1340',\n",
" '1341',\n",
" '1342',\n",
" '1402',\n",
" '1409',\n",
" '1410',\n",
" '1413',\n",
" '1414',\n",
" '1416',\n",
" '1417',\n",
" '1418',\n",
" '1419',\n",
" '1423',\n",
" '1432',\n",
" '1434',\n",
" '1435',\n",
" '1436',\n",
" '1437',\n",
" '1438',\n",
" '1439',\n",
" '1440',\n",
" '1441',\n",
" '1442',\n",
" '1443',\n",
" '1444',\n",
" '1445',\n",
" '1446',\n",
" '1447',\n",
" '1449',\n",
" '1451',\n",
" '1452',\n",
" '1453',\n",
" '1454',\n",
" '1455',\n",
" '1456',\n",
" '1457',\n",
" '1459',\n",
" '1460',\n",
" '1463',\n",
" '1464',\n",
" '1465',\n",
" '1466',\n",
" '1467',\n",
" '1468',\n",
" '1470',\n",
" '1471',\n",
" '1472',\n",
" '1473',\n",
" '1474',\n",
" '1475',\n",
" '1476',\n",
" '1477',\n",
" '1503',\n",
" '1504',\n",
" '1506',\n",
" '1507',\n",
" '1512',\n",
" '1513',\n",
" '1514',\n",
" '1515',\n",
" '1516',\n",
" '1517',\n",
" '1519',\n",
" '1521',\n",
" '1522',\n",
" '1524',\n",
" '1525',\n",
" '1526',\n",
" '1527',\n",
" '1528',\n",
" '1529',\n",
" '1530',\n",
" '1531',\n",
" '1532',\n",
" '1533',\n",
" '1535',\n",
" '1536',\n",
" '1537',\n",
" '1538',\n",
" '1539',\n",
" '1540',\n",
" '1541',\n",
" '1558',\n",
" '1560',\n",
" '1568',\n",
" '1582',\n",
" '1583',\n",
" '1587',\n",
" '1589',\n",
" '1590',\n",
" '1597',\n",
" '1598',\n",
" '1603',\n",
" '1604',\n",
" '1605',\n",
" '1608',\n",
" '1609',\n",
" '1611',\n",
" '1612',\n",
" '1614',\n",
" '1615',\n",
" '1616',\n",
" '1617',\n",
" '1618',\n",
" '1626',\n",
" '1701',\n",
" '1702',\n",
" '1707',\n",
" '1708',\n",
" '1709',\n",
" '1710',\n",
" '1711',\n",
" '1712',\n",
" '1713',\n",
" '1714',\n",
" '1717',\n",
" '1718',\n",
" '1720',\n",
" '1721',\n",
" '1722',\n",
" '1723',\n",
" '1725',\n",
" '1726',\n",
" '1727',\n",
" '1730',\n",
" '1731',\n",
" '1732',\n",
" '1733',\n",
" '1734',\n",
" '1735',\n",
" '1736',\n",
" '1737',\n",
" '1752',\n",
" '1760',\n",
" '1762',\n",
" '1773',\n",
" '1776',\n",
" '1783',\n",
" '1786',\n",
" '1789',\n",
" '1795',\n",
" '1802',\n",
" '1805',\n",
" '1806',\n",
" '1808',\n",
" '1809',\n",
" '1810',\n",
" '1817',\n",
" '1903',\n",
" '1904',\n",
" '1905',\n",
" '1906',\n",
" '1907',\n",
" '1909',\n",
" '2002',\n",
" '2006',\n",
" '2007',\n",
" '2008',\n",
" '2009',\n",
" '2010',\n",
" '2012',\n",
" '2013',\n",
" '2014',\n",
" '2015',\n",
" '2017',\n",
" '2020',\n",
" '2022',\n",
" '2023',\n",
" '2024',\n",
" '2025',\n",
" '2027',\n",
" '2028',\n",
" '2029',\n",
" '2030',\n",
" '2031',\n",
" '2032',\n",
" '2033',\n",
" '2034',\n",
" '2038',\n",
" '2049',\n",
" '2059',\n",
" '2062',\n",
" '2069',\n",
" '2101',\n",
" '2102',\n",
" '2103',\n",
" '2104',\n",
" '2105',\n",
" '2106',\n",
" '2107',\n",
" '2108',\n",
" '2109',\n",
" '2114',\n",
" '2115',\n",
" '2201',\n",
" '2204',\n",
" '2206',\n",
" '2207',\n",
" '2208',\n",
" '2211',\n",
" '2227',\n",
" '2228',\n",
" '2231',\n",
" '2233',\n",
" '2236',\n",
" '2239',\n",
" '2241',\n",
" '2243',\n",
" '2247',\n",
" '2250',\n",
" '2301',\n",
" '2302',\n",
" '2303',\n",
" '2305',\n",
" '2308',\n",
" '2312',\n",
" '2313',\n",
" '2314',\n",
" '2316',\n",
" '2317',\n",
" '2321',\n",
" '2323',\n",
" '2324',\n",
" '2327',\n",
" '2328',\n",
" '2329',\n",
" '2330',\n",
" '2331',\n",
" '2332',\n",
" '2337',\n",
" '2338',\n",
" '2340',\n",
" '2342',\n",
" '2344',\n",
" '2345',\n",
" '2347',\n",
" '2348',\n",
" '2349',\n",
" '2351',\n",
" '2352',\n",
" '2353',\n",
" '2354',\n",
" '2355',\n",
" '2356',\n",
" '2357',\n",
" '2358',\n",
" '2359',\n",
" '2360',\n",
" '2362',\n",
" '2363',\n",
" '2364',\n",
" '2365',\n",
" '2367',\n",
" '2368',\n",
" '2369',\n",
" '2371',\n",
" '2373',\n",
" '2374',\n",
" '2375',\n",
" '2376',\n",
" '2377',\n",
" '2379',\n",
" '2380',\n",
" '2382',\n",
" '2383',\n",
" '2385',\n",
" '2387',\n",
" '2388',\n",
" '2390',\n",
" '2392',\n",
" '2393',\n",
" '2395',\n",
" '2397',\n",
" '2399',\n",
" '2401',\n",
" '2402',\n",
" '2404',\n",
" '2405',\n",
" '2406',\n",
" '2408',\n",
" '2409',\n",
" '2412',\n",
" '2413',\n",
" '2414',\n",
" '2415',\n",
" '2417',\n",
" '2419',\n",
" '2420',\n",
" '2421',\n",
" '2423',\n",
" '2424',\n",
" '2425',\n",
" '2426',\n",
" '2427',\n",
" '2428',\n",
" '2429',\n",
" '2430',\n",
" '2431',\n",
" '2433',\n",
" '2434',\n",
" '2436',\n",
" '2438',\n",
" '2439',\n",
" '2440',\n",
" '2441',\n",
" '2442',\n",
" '2443',\n",
" '2444',\n",
" '2449',\n",
" '2450',\n",
" '2451',\n",
" '2453',\n",
" '2454',\n",
" '2455',\n",
" '2457',\n",
" '2458',\n",
" '2459',\n",
" '2460',\n",
" '2461',\n",
" '2462',\n",
" '2464',\n",
" '2465',\n",
" '2466',\n",
" '2467',\n",
" '2468',\n",
" '2471',\n",
" '2472',\n",
" '2474',\n",
" '2476',\n",
" '2477',\n",
" '2478',\n",
" '2480',\n",
" '2481',\n",
" '2482',\n",
" '2483',\n",
" '2484',\n",
" '2485',\n",
" '2486',\n",
" '2488',\n",
" '2489',\n",
" '2491',\n",
" '2492',\n",
" '2493',\n",
" '2495',\n",
" '2496',\n",
" '2497',\n",
" '2498',\n",
" '2501',\n",
" '2504',\n",
" '2505',\n",
" '2506',\n",
" '2509',\n",
" '2511',\n",
" '2514',\n",
" '2515',\n",
" '2516',\n",
" '2520',\n",
" '2524',\n",
" '2527',\n",
" '2528',\n",
" '2530',\n",
" '2534',\n",
" '2535',\n",
" '2536',\n",
" '2537',\n",
" '2538',\n",
" '2539',\n",
" '2540',\n",
" '2542',\n",
" '2543',\n",
" '2545',\n",
" '2546',\n",
" '2547',\n",
" '2548',\n",
" '2597',\n",
" '2601',\n",
" '2603',\n",
" '2605',\n",
" '2606',\n",
" '2607',\n",
" '2608',\n",
" '2609',\n",
" '2610',\n",
" '2611',\n",
" '2612',\n",
" '2613',\n",
" '2614',\n",
" '2615',\n",
" '2616',\n",
" '2617',\n",
" '2618',\n",
" '2630',\n",
" '2633',\n",
" '2634',\n",
" '2636',\n",
" '2637',\n",
" '2642',\n",
" '2701',\n",
" '2702',\n",
" '2704',\n",
" '2705',\n",
" '2706',\n",
" '2707',\n",
" '2712',\n",
" '2722',\n",
" '2723',\n",
" '2727',\n",
" '2731',\n",
" '2739',\n",
" '2748',\n",
" '2753',\n",
" '2801',\n",
" '2809',\n",
" '2812',\n",
" '2816',\n",
" '2820',\n",
" '2832',\n",
" '2834',\n",
" '2836',\n",
" '2838',\n",
" '2841',\n",
" '2845',\n",
" '2849',\n",
" '2850',\n",
" '2851',\n",
" '2852',\n",
" '2855',\n",
" '2867',\n",
" '2880',\n",
" '2881',\n",
" '2882',\n",
" '2883',\n",
" '2884',\n",
" '2885',\n",
" '2886',\n",
" '2887',\n",
" '2888',\n",
" '2889',\n",
" '2890',\n",
" '2891',\n",
" '2892',\n",
" '2897',\n",
" '2901',\n",
" '2903',\n",
" '2904',\n",
" '2905',\n",
" '2906',\n",
" '2908',\n",
" '2910',\n",
" '2911',\n",
" '2912',\n",
" '2913',\n",
" '2915',\n",
" '2923',\n",
" '2929',\n",
" '2936',\n",
" '2939',\n",
" '2945',\n",
" '3002',\n",
" '3003',\n",
" '3004',\n",
" '3005',\n",
" '3006',\n",
" '3008',\n",
" '3010',\n",
" '3011',\n",
" '3013',\n",
" '3014',\n",
" '3015',\n",
" '3016',\n",
" '3017',\n",
" '3018',\n",
" '3019',\n",
" '3021',\n",
" '3022',\n",
" '3023',\n",
" '3024',\n",
" '3025',\n",
" '3026',\n",
" '3027',\n",
" '3028',\n",
" '3029',\n",
" '3030',\n",
" '3031',\n",
" '3032',\n",
" '3033',\n",
" '3034',\n",
" '3035',\n",
" '3036',\n",
" '3037',\n",
" '3038',\n",
" '3040',\n",
" '3041',\n",
" '3042',\n",
" '3043',\n",
" '3044',\n",
" '3045',\n",
" '3046',\n",
" '3047',\n",
" '3048',\n",
" '3049',\n",
" '3050',\n",
" '3051',\n",
" '3052',\n",
" '3054',\n",
" '3055',\n",
" '3056',\n",
" '3057',\n",
" '3058',\n",
" '3059',\n",
" '3060',\n",
" '3062',\n",
" '3090',\n",
" '3092',\n",
" '3094',\n",
" '3130',\n",
" '3138',\n",
" '3149',\n",
" '3164',\n",
" '3167',\n",
" '3189',\n",
" '3209',\n",
" '3229',\n",
" '3231',\n",
" '3257',\n",
" '3266',\n",
" '3296',\n",
" '3305',\n",
" '3308',\n",
" '3311',\n",
" '3312',\n",
" '3321',\n",
" '3338',\n",
" '3346',\n",
" '3356',\n",
" '3376',\n",
" '3380',\n",
" '3383',\n",
" '3406',\n",
" '3413',\n",
" '3416',\n",
" '3419',\n",
" '3432',\n",
" '3437',\n",
" '3443',\n",
" '3450',\n",
" '3454',\n",
" '3481',\n",
" '3494',\n",
" '3501',\n",
" '3504',\n",
" '3515',\n",
" '3518',\n",
" '3528',\n",
" '3530',\n",
" '3532',\n",
" '3533',\n",
" '3535',\n",
" '3536',\n",
" '3543',\n",
" '3545',\n",
" '3550',\n",
" '3557',\n",
" '3563',\n",
" '3576',\n",
" '3583',\n",
" '3588',\n",
" '3591',\n",
" '3592',\n",
" '3593',\n",
" '3596',\n",
" '3605',\n",
" '3607',\n",
" '3617',\n",
" '3622',\n",
" '3645',\n",
" '3653',\n",
" '3661',\n",
" '3665',\n",
" '3669',\n",
" '3673',\n",
" '3679',\n",
" '3682',\n",
" '3686',\n",
" '3694',\n",
" '3701',\n",
" '3702',\n",
" '3703',\n",
" '3704',\n",
" '3705',\n",
" '3706',\n",
" '3708',\n",
" '3711',\n",
" '3712',\n",
" '3714',\n",
" '4104',\n",
" '4106',\n",
" '4108',\n",
" '4119',\n",
" '4133',\n",
" '4137',\n",
" '4141',\n",
" '4142',\n",
" '4148',\n",
" '4155',\n",
" '4164',\n",
" '4190',\n",
" '4306',\n",
" '4414',\n",
" '4426',\n",
" '4438',\n",
" '4439',\n",
" '4440',\n",
" '4526',\n",
" '4532',\n",
" '4536',\n",
" '4540',\n",
" '4545',\n",
" '4551',\n",
" '4552',\n",
" '4555',\n",
" '4557',\n",
" '4560',\n",
" '4562',\n",
" '4564',\n",
" '4566',\n",
" '4571',\n",
" '4572',\n",
" '4576',\n",
" '4581',\n",
" '4720',\n",
" '4722',\n",
" '4737',\n",
" '4739',\n",
" '4746',\n",
" '4755',\n",
" '4763',\n",
" '4764',\n",
" '4766',\n",
" '4770',\n",
" '4807',\n",
" '4904',\n",
" '4906',\n",
" '4912',\n",
" '4915',\n",
" '4916',\n",
" '4919',\n",
" '4927',\n",
" '4930',\n",
" '4934',\n",
" '4935',\n",
" '4938',\n",
" '4942',\n",
" '4943',\n",
" '4952',\n",
" '4956',\n",
" '4958',\n",
" '4960',\n",
" '4961',\n",
" '4967',\n",
" '4968',\n",
" '4976',\n",
" '4977',\n",
" '4989',\n",
" '4994',\n",
" '4999',\n",
" '5007',\n",
" '5203',\n",
" '5215',\n",
" '5222',\n",
" '5225',\n",
" '5234',\n",
" '5243',\n",
" '5244',\n",
" '5258',\n",
" '5269',\n",
" '5283',\n",
" '5284',\n",
" '5285',\n",
" '5288',\n",
" '5306',\n",
" '5388',\n",
" '5434',\n",
" '5469',\n",
" '5471',\n",
" '5484',\n",
" '5515',\n",
" '5519',\n",
" '5521',\n",
" '5522',\n",
" '5525',\n",
" '5531',\n",
" '5533',\n",
" '5534',\n",
" '5538',\n",
" '5546',\n",
" '5607',\n",
" '5608',\n",
" '5706',\n",
" '5871',\n",
" '5876',\n",
" '5880',\n",
" '5906',\n",
" '5907',\n",
" '6005',\n",
" '6024',\n",
" '6108',\n",
" '6112',\n",
" '6115',\n",
" '6116',\n",
" '6117',\n",
" '6120',\n",
" '6128',\n",
" '6133',\n",
" '6136',\n",
" '6139',\n",
" '6141',\n",
" '6142',\n",
" '6152',\n",
" '6153',\n",
" '6155',\n",
" '6164',\n",
" '6165',\n",
" '6166',\n",
" '6168',\n",
" '6172',\n",
" '6176',\n",
" '6177',\n",
" '6183',\n",
" '6184',\n",
" '6189',\n",
" '6191',\n",
" '6192',\n",
" '6196',\n",
" '6197',\n",
" '6201',\n",
" '6202',\n",
" '6205',\n",
" '6206',\n",
" '6209',\n",
" '6213',\n",
" '6214',\n",
" '6215',\n",
" '6216',\n",
" '6224',\n",
" '6225',\n",
" '6226',\n",
" '6230',\n",
" '6235',\n",
" '6239',\n",
" '6243',\n",
" '6251',\n",
" '6257',\n",
" '6269',\n",
" '6271',\n",
" '6277',\n",
" '6278',\n",
" '6281',\n",
" '6282',\n",
" '6283',\n",
" '6285',\n",
" '6288',\n",
" '6289',\n",
" '6405',\n",
" '6409',\n",
" '6412',\n",
" '6414',\n",
" '6415',\n",
" '6416',\n",
" '6426',\n",
" '6431',\n",
" '6438',\n",
" '6442',\n",
" '6443',\n",
" '6449',\n",
" '6451',\n",
" '6456',\n",
" '6464',\n",
" '6477',\n",
" '6491',\n",
" '6504',\n",
" '6505',\n",
" '6515',\n",
" '6525',\n",
" '6531',\n",
" '6533',\n",
" '6541',\n",
" '6552',\n",
" '6558',\n",
" '6573',\n",
" '6579',\n",
" '6581',\n",
" '6582',\n",
" '6591',\n",
" '6592',\n",
" '6598',\n",
" '6605',\n",
" '6625',\n",
" '6641',\n",
" '6655',\n",
" '6666',\n",
" '6668',\n",
" '6669',\n",
" '6670',\n",
" '6671',\n",
" '6672',\n",
" '6674',\n",
" '6691',\n",
" '6698',\n",
" '6706',\n",
" '6715',\n",
" '6719',\n",
" '6743',\n",
" '6754',\n",
" '6756',\n",
" '6768',\n",
" '6770',\n",
" '6776',\n",
" '6781',\n",
" '6790',\n",
" '6792',\n",
" '6806',\n",
" '8011',\n",
" '8016',\n",
" '8021',\n",
" '8028',\n",
" '8033',\n",
" '8039',\n",
" '8046',\n",
" '8070',\n",
" '8072',\n",
" '8081',\n",
" '8101',\n",
" '8103',\n",
" '8104',\n",
" '8105',\n",
" '8110',\n",
" '8112',\n",
" '8114',\n",
" '8131',\n",
" '8150',\n",
" '8163',\n",
" '8201',\n",
" '8210',\n",
" '8213',\n",
" '8215',\n",
" '8222',\n",
" '8249',\n",
" '8261',\n",
" '8271',\n",
" '8341',\n",
" '8367',\n",
" '8374',\n",
" '8404',\n",
" '8411',\n",
" '8422',\n",
" '8429',\n",
" '8438',\n",
" '8442',\n",
" '8443',\n",
" '8454',\n",
" '8462',\n",
" '8463',\n",
" '8464',\n",
" '8466',\n",
" '8467',\n",
" '8473',\n",
" '8478',\n",
" '8480',\n",
" '8481',\n",
" '8482',\n",
" '8488',\n",
" '8499',\n",
" '8926',\n",
" '8940',\n",
" '8996',\n",
" '9802',\n",
" '9902',\n",
" '9904',\n",
" '9905',\n",
" '9906',\n",
" '9907',\n",
" '9908',\n",
" '9910',\n",
" '9911',\n",
" '9912',\n",
" '9914',\n",
" '9917',\n",
" '9918',\n",
" '9919',\n",
" '9921',\n",
" '9924',\n",
" '9925',\n",
" '9926',\n",
" '9927',\n",
" '9928',\n",
" '9929',\n",
" '9930',\n",
" '9931',\n",
" '9933',\n",
" '9934',\n",
" '9935',\n",
" '9937',\n",
" '9938',\n",
" '9939',\n",
" '9940',\n",
" '9941',\n",
" '9942',\n",
" '9943',\n",
" '9944',\n",
" '9945',\n",
" '9946',\n",
" '9955',\n",
" '9958']"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"sec_code = tejapi.get('TWN/EWNPRCSTD', chinese_column_name = True)\n",
"\n",
"condition = (sec_code['上市別'] == 'TSE') & (sec_code['證券種類名稱'] == '普通股')\n",
"pub_common_stk = sec_code.loc[condition, '證券碼'].to_list()\n",
"pub_common_stk"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "1c99d781",
"metadata": {},
"outputs": [],
"source": [
"fin_data = tejapi.get('TWN/EWIFINQ',\n",
" coid = pub_common_stk, \n",
" mdate= {'gte': '2017-01-01', 'lte': '2021-12-31'},\n",
" opts={'columns': ['coid', 'mdate', 'ac_r103', 'ac_r403']},\n",
" paginate = True,\n",
" chinese_column_name = True\n",
" )"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "531f456d",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>證券碼</th>\n",
" <th>財務資料日</th>\n",
" <th>ROE(A)-稅後</th>\n",
" <th>營業利益成長率</th>\n",
" </tr>\n",
" <tr>\n",
" <th>None</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1101</td>\n",
" <td>2019-06-01</td>\n",
" <td>14.96</td>\n",
" <td>-13.75</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1101</td>\n",
" <td>2020-06-01</td>\n",
" <td>17.12</td>\n",
" <td>20.40</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1101</td>\n",
" <td>2017-06-01</td>\n",
" <td>8.40</td>\n",
" <td>7.69</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>1101</td>\n",
" <td>2020-09-01</td>\n",
" <td>17.00</td>\n",
" <td>11.52</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1101</td>\n",
" <td>2019-03-01</td>\n",
" <td>8.32</td>\n",
" <td>8.22</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18971</th>\n",
" <td>9958</td>\n",
" <td>2018-09-01</td>\n",
" <td>0.72</td>\n",
" <td>137.59</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18972</th>\n",
" <td>9958</td>\n",
" <td>2021-09-01</td>\n",
" <td>20.08</td>\n",
" <td>30.52</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18973</th>\n",
" <td>9958</td>\n",
" <td>2018-06-01</td>\n",
" <td>0.56</td>\n",
" <td>72.61</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18974</th>\n",
" <td>9958</td>\n",
" <td>2019-03-01</td>\n",
" <td>0.52</td>\n",
" <td>25.10</td>\n",
" </tr>\n",
" <tr>\n",
" <th>18975</th>\n",
" <td>9958</td>\n",
" <td>2020-12-01</td>\n",
" <td>5.28</td>\n",
" <td>-16.39</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>18160 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" 證券碼 財務資料日 ROE(A)-稅後 營業利益成長率\n",
"None \n",
"0 1101 2019-06-01 14.96 -13.75\n",
"1 1101 2020-06-01 17.12 20.40\n",
"2 1101 2017-06-01 8.40 7.69\n",
"3 1101 2020-09-01 17.00 11.52\n",
"4 1101 2019-03-01 8.32 8.22\n",
"... ... ... ... ...\n",
"18971 9958 2018-09-01 0.72 137.59\n",
"18972 9958 2021-09-01 20.08 30.52\n",
"18973 9958 2018-06-01 0.56 72.61\n",
"18974 9958 2019-03-01 0.52 25.10\n",
"18975 9958 2020-12-01 5.28 -16.39\n",
"\n",
"[18160 rows x 4 columns]"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"fin_data = fin_data.dropna()\n",
"fin_data"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "a6cc03a5",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>證券代碼</th>\n",
" <th>年月日</th>\n",
" <th>收盤價(元)</th>\n",
" </tr>\n",
" <tr>\n",
" <th>None</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1101</td>\n",
" <td>2017-01-03</td>\n",
" <td>35.15</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1101</td>\n",
" <td>2017-01-04</td>\n",
" <td>35.25</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1101</td>\n",
" <td>2017-01-05</td>\n",
" <td>35.25</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>1101</td>\n",
" <td>2017-01-06</td>\n",
" <td>35.25</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1101</td>\n",
" <td>2017-01-09</td>\n",
" <td>34.90</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1219</th>\n",
" <td>9958</td>\n",
" <td>2021-12-24</td>\n",
" <td>110.50</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1220</th>\n",
" <td>9958</td>\n",
" <td>2021-12-27</td>\n",
" <td>110.50</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1221</th>\n",
" <td>9958</td>\n",
" <td>2021-12-28</td>\n",
" <td>110.50</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1222</th>\n",
" <td>9958</td>\n",
" <td>2021-12-29</td>\n",
" <td>111.00</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1223</th>\n",
" <td>9958</td>\n",
" <td>2021-12-30</td>\n",
" <td>110.50</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>1121251 rows × 3 columns</p>\n",
"</div>"
],
"text/plain": [
" 證券代碼 年月日 收盤價(元)\n",
"None \n",
"0 1101 2017-01-03 35.15\n",
"1 1101 2017-01-04 35.25\n",
"2 1101 2017-01-05 35.25\n",
"3 1101 2017-01-06 35.25\n",
"4 1101 2017-01-09 34.90\n",
"... ... ... ...\n",
"1219 9958 2021-12-24 110.50\n",
"1220 9958 2021-12-27 110.50\n",
"1221 9958 2021-12-28 110.50\n",
"1222 9958 2021-12-29 111.00\n",
"1223 9958 2021-12-30 110.50\n",
"\n",
"[1121251 rows x 3 columns]"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ret_df = pd.DataFrame()\n",
"for i in pub_common_stk:\n",
" ret_data = tejapi.get('TWN/APRCD',\n",
" coid = i, \n",
" mdate= {'gte': '2017-01-01', 'lte': '2021-12-31'},\n",
" opts={'columns': ['coid', 'mdate', 'close_d']},\n",
" paginate = True,\n",
" chinese_column_name = True\n",
" )\n",
" ret_df = pd.concat([ret_df, ret_data], axis = 0)\n",
" \n",
"ret_df"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "f521e222",
"metadata": {},
"outputs": [],
"source": [
"ret_df_copy = ret_df.copy()"
]
},
{
"cell_type": "markdown",
"id": "2645d41c",
"metadata": {},
"source": [
"## Data Pre-processing"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "ab378b0d",
"metadata": {},
"outputs": [],
"source": [
"def compare(df1, df2):\n",
" df2 = df2.rename(columns = {'證券碼':'證券代碼', '財務資料日':'年月日'})\n",
" compare = pd.merge(df1, df2, how='inner', on=['證券代碼', '年月日'])\n",
" result1 = pd.concat([compare['年月日'], compare['證券代碼']], axis = 1)\n",
" result2 = pd.merge(df1, result1, how='inner', on =['證券代碼', '年月日'])\n",
" return result2"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "2dc2697f",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>年月日</th>\n",
" <th>收盤價(元)</th>\n",
" </tr>\n",
" <tr>\n",
" <th>證券代碼</th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>1101</th>\n",
" <td>2017-03-01</td>\n",
" <td>37.70</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1101</th>\n",
" <td>2017-06-01</td>\n",
" <td>34.35</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1101</th>\n",
" <td>2017-09-01</td>\n",
" <td>34.65</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1101</th>\n",
" <td>2017-12-01</td>\n",
" <td>33.70</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1101</th>\n",
" <td>2018-03-01</td>\n",
" <td>37.00</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9958</th>\n",
" <td>2020-09-01</td>\n",
" <td>120.00</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9958</th>\n",
" <td>2020-12-01</td>\n",
" <td>128.50</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9958</th>\n",
" <td>2021-06-01</td>\n",
" <td>113.50</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9958</th>\n",
" <td>2021-09-01</td>\n",
" <td>111.00</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9958</th>\n",
" <td>2021-12-01</td>\n",
" <td>112.00</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>10656 rows × 2 columns</p>\n",
"</div>"
],
"text/plain": [
" 年月日 收盤價(元)\n",
"證券代碼 \n",
"1101 2017-03-01 37.70\n",
"1101 2017-06-01 34.35\n",
"1101 2017-09-01 34.65\n",
"1101 2017-12-01 33.70\n",
"1101 2018-03-01 37.00\n",
"... ... ...\n",
"9958 2020-09-01 120.00\n",
"9958 2020-12-01 128.50\n",
"9958 2021-06-01 113.50\n",
"9958 2021-09-01 111.00\n",
"9958 2021-12-01 112.00\n",
"\n",
"[10656 rows x 2 columns]"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ret_df2 = compare(ret_df, fin_data)\n",
"ret_df2 = ret_df2.set_index(['證券代碼'])\n",
"ret_df2"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "aa018b9a",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"<ipython-input-9-a7dbca6471af>:1: DeprecationWarning: The default dtype for empty Series will be 'object' instead of 'float64' in a future version. Specify a dtype explicitly to silence this warning.\n",
" ret_df2['報酬%'] = pd.Series()\n",
"<ipython-input-9-a7dbca6471af>:3: SettingWithCopyWarning: \n",
"A value is trying to be set on a copy of a slice from a DataFrame.\n",
"Try using .loc[row_indexer,col_indexer] = value instead\n",
"\n",
"See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
" ret_df2.loc[i]['報酬%'] = pd.Series(ret_df2.loc[i]['收盤價(元)']).pct_change(1)*100\n",
"<ipython-input-9-a7dbca6471af>:3: SettingWithCopyWarning: \n",
"A value is trying to be set on a copy of a slice from a DataFrame\n",
"\n",
"See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
" ret_df2.loc[i]['報酬%'] = pd.Series(ret_df2.loc[i]['收盤價(元)']).pct_change(1)*100\n"
]
}
],
"source": [
"ret_df2['報酬%'] = pd.Series()\n",
"for i in ret_df2.index.values:\n",
" ret_df2.loc[i]['報酬%'] = pd.Series(ret_df2.loc[i]['收盤價(元)']).pct_change(1)*100\n",
"ret_df2 = ret_df2.dropna().reset_index()"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "f65eeef4",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>證券代碼</th>\n",
" <th>年月日</th>\n",
" <th>收盤價(元)</th>\n",
" <th>報酬%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>1101</td>\n",
" <td>2017-06-01</td>\n",
" <td>34.35</td>\n",
" <td>-8.885942</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>1101</td>\n",
" <td>2017-09-01</td>\n",
" <td>34.65</td>\n",
" <td>0.873362</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>1101</td>\n",
" <td>2017-12-01</td>\n",
" <td>33.70</td>\n",
" <td>-2.741703</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>1101</td>\n",
" <td>2018-03-01</td>\n",
" <td>37.00</td>\n",
" <td>9.792285</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1101</td>\n",
" <td>2018-06-01</td>\n",
" <td>43.75</td>\n",
" <td>18.243243</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9721</th>\n",
" <td>9958</td>\n",
" <td>2020-09-01</td>\n",
" <td>120.00</td>\n",
" <td>31.578947</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9722</th>\n",
" <td>9958</td>\n",
" <td>2020-12-01</td>\n",
" <td>128.50</td>\n",
" <td>7.083333</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9723</th>\n",
" <td>9958</td>\n",
" <td>2021-06-01</td>\n",
" <td>113.50</td>\n",
" <td>-11.673152</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9724</th>\n",
" <td>9958</td>\n",
" <td>2021-09-01</td>\n",
" <td>111.00</td>\n",
" <td>-2.202643</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9725</th>\n",
" <td>9958</td>\n",
" <td>2021-12-01</td>\n",
" <td>112.00</td>\n",
" <td>0.900901</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>9726 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" 證券代碼 年月日 收盤價(元) 報酬%\n",
"0 1101 2017-06-01 34.35 -8.885942\n",
"1 1101 2017-09-01 34.65 0.873362\n",
"2 1101 2017-12-01 33.70 -2.741703\n",
"3 1101 2018-03-01 37.00 9.792285\n",
"4 1101 2018-06-01 43.75 18.243243\n",
"... ... ... ... ...\n",
"9721 9958 2020-09-01 120.00 31.578947\n",
"9722 9958 2020-12-01 128.50 7.083333\n",
"9723 9958 2021-06-01 113.50 -11.673152\n",
"9724 9958 2021-09-01 111.00 -2.202643\n",
"9725 9958 2021-12-01 112.00 0.900901\n",
"\n",
"[9726 rows x 4 columns]"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ret_df2"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "64f18a76",
"metadata": {},
"outputs": [],
"source": [
"def compare2(df1, df2):\n",
" df2 = df2.rename(columns = {'證券碼':'證券代碼', '財務資料日':'年月日'})\n",
" compare = pd.merge(df1, df2, how='inner', on=['證券代碼', '年月日'])\n",
" return compare"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "d02390f2",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th></th>\n",
" <th>收盤價(元)</th>\n",
" <th>報酬%</th>\n",
" <th>ROE(A)-稅後</th>\n",
" <th>營業利益成長率</th>\n",
" </tr>\n",
" <tr>\n",
" <th>證券代碼</th>\n",
" <th>年月日</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">1101</th>\n",
" <th>2017-06-01</th>\n",
" <td>34.35</td>\n",
" <td>-8.885942</td>\n",
" <td>8.40</td>\n",
" <td>7.69</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-09-01</th>\n",
" <td>34.65</td>\n",
" <td>0.873362</td>\n",
" <td>7.20</td>\n",
" <td>-14.14</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-12-01</th>\n",
" <td>33.70</td>\n",
" <td>-2.741703</td>\n",
" <td>8.92</td>\n",
" <td>-5.69</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2018-03-01</th>\n",
" <td>37.00</td>\n",
" <td>9.792285</td>\n",
" <td>7.20</td>\n",
" <td>139.53</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2018-06-01</th>\n",
" <td>43.75</td>\n",
" <td>18.243243</td>\n",
" <td>18.60</td>\n",
" <td>121.12</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">9958</th>\n",
" <th>2020-09-01</th>\n",
" <td>120.00</td>\n",
" <td>31.578947</td>\n",
" <td>22.92</td>\n",
" <td>716.83</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2020-12-01</th>\n",
" <td>128.50</td>\n",
" <td>7.083333</td>\n",
" <td>5.28</td>\n",
" <td>-16.39</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-06-01</th>\n",
" <td>113.50</td>\n",
" <td>-11.673152</td>\n",
" <td>21.08</td>\n",
" <td>43.11</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-09-01</th>\n",
" <td>111.00</td>\n",
" <td>-2.202643</td>\n",
" <td>20.08</td>\n",
" <td>30.52</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-12-01</th>\n",
" <td>112.00</td>\n",
" <td>0.900901</td>\n",
" <td>6.96</td>\n",
" <td>59.19</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>9726 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" 收盤價(元) 報酬% ROE(A)-稅後 營業利益成長率\n",
"證券代碼 年月日 \n",
"1101 2017-06-01 34.35 -8.885942 8.40 7.69\n",
" 2017-09-01 34.65 0.873362 7.20 -14.14\n",
" 2017-12-01 33.70 -2.741703 8.92 -5.69\n",
" 2018-03-01 37.00 9.792285 7.20 139.53\n",
" 2018-06-01 43.75 18.243243 18.60 121.12\n",
"... ... ... ... ...\n",
"9958 2020-09-01 120.00 31.578947 22.92 716.83\n",
" 2020-12-01 128.50 7.083333 5.28 -16.39\n",
" 2021-06-01 113.50 -11.673152 21.08 43.11\n",
" 2021-09-01 111.00 -2.202643 20.08 30.52\n",
" 2021-12-01 112.00 0.900901 6.96 59.19\n",
"\n",
"[9726 rows x 4 columns]"
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data = compare2(ret_df2, fin_data).set_index(['證券代碼','年月日'])\n",
"data"
]
},
{
"cell_type": "markdown",
"id": "ac94fdda",
"metadata": {},
"source": [
"## Outlier"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "2ab34f41",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"Text(0, 0.5, '營業利益成長率')"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x576 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.rcParams['font.sans-serif'] = ['Arial Unicode MS']\n",
"\n",
"plt.figure(figsize=(15,8))\n",
"\n",
"plt.scatter(data['ROE(A)-稅後'], data['營業利益成長率'])\n",
"plt.xlabel('股東權益報酬率')\n",
"plt.ylabel('營業利益成長率')"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "505751cf",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(array([ 39., 39., 62., 69., 70., 82., 111., 130., 128., 183., 206.,\n",
" 203., 230., 257., 314., 365., 423., 447., 472., 452., 403., 346.,\n",
" 306., 261., 236., 195., 167., 172., 148., 142., 120., 124., 97.,\n",
" 107., 83., 80., 82., 65., 75., 57., 49., 56., 51., 46.,\n",
" 44., 35., 38., 36., 36., 23., 27., 39., 23., 20., 22.,\n",
" 17., 15., 18., 12., 22., 14., 19., 15., 13., 22., 6.,\n",
" 16., 13., 9., 10., 17., 8., 4., 11., 8., 11., 7.,\n",
" 13., 6., 5., 7., 12., 8., 10., 3., 5., 5., 7.,\n",
" 7., 2., 3., 10., 2., 5., 4., 6., 6., 1., 7.,\n",
" 7.]),\n",
" array([-142.28 , -134.5593, -126.8386, -119.1179, -111.3972, -103.6765,\n",
" -95.9558, -88.2351, -80.5144, -72.7937, -65.073 , -57.3523,\n",
" -49.6316, -41.9109, -34.1902, -26.4695, -18.7488, -11.0281,\n",
" -3.3074, 4.4133, 12.134 , 19.8547, 27.5754, 35.2961,\n",
" 43.0168, 50.7375, 58.4582, 66.1789, 73.8996, 81.6203,\n",
" 89.341 , 97.0617, 104.7824, 112.5031, 120.2238, 127.9445,\n",
" 135.6652, 143.3859, 151.1066, 158.8273, 166.548 , 174.2687,\n",
" 181.9894, 189.7101, 197.4308, 205.1515, 212.8722, 220.5929,\n",
" 228.3136, 236.0343, 243.755 , 251.4757, 259.1964, 266.9171,\n",
" 274.6378, 282.3585, 290.0792, 297.7999, 305.5206, 313.2413,\n",
" 320.962 , 328.6827, 336.4034, 344.1241, 351.8448, 359.5655,\n",
" 367.2862, 375.0069, 382.7276, 390.4483, 398.169 , 405.8897,\n",
" 413.6104, 421.3311, 429.0518, 436.7725, 444.4932, 452.2139,\n",
" 459.9346, 467.6553, 475.376 , 483.0967, 490.8174, 498.5381,\n",
" 506.2588, 513.9795, 521.7002, 529.4209, 537.1416, 544.8623,\n",
" 552.583 , 560.3037, 568.0244, 575.7451, 583.4658, 591.1865,\n",
" 598.9072, 606.6279, 614.3486, 622.0693, 629.79 ]),\n",
" <BarContainer object of 100 artists>)"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x576 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"def remove_outlier(data, ntsd):\n",
" ub = data.mean() + ntsd * data.std()\n",
" lb = data.mean() - ntsd * data.std()\n",
"\n",
" return (data > lb) & (data < ub)\n",
"\n",
"no_out = remove_outlier(data['ROE(A)-稅後'], 2) & remove_outlier(data['營業利益成長率'], 0.05)\n",
"data_no = data[no_out].dropna()\n",
"\n",
"plt.figure(figsize=(15,8))\n",
"plt.hist(data_no['ROE(A)-稅後'], bins = 100)\n",
"plt.hist(data_no['營業利益成長率'], bins = 100, alpha =0.7)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "132c64a7",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.collections.PathCollection at 0x7fecec9a2a00>"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x576 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.rcParams['font.sans-serif'] = ['Arial Unicode MS']\n",
"\n",
"plt.figure(figsize=(15,8))\n",
"\n",
"plt.scatter(data_no['ROE(A)-稅後'], data_no['營業利益成長率'])"
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "81fa7215",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th></th>\n",
" <th>收盤價(元)</th>\n",
" <th>報酬%</th>\n",
" <th>ROE(A)-稅後</th>\n",
" <th>營業利益成長率</th>\n",
" </tr>\n",
" <tr>\n",
" <th>證券代碼</th>\n",
" <th>年月日</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">1101</th>\n",
" <th>2017-06-01</th>\n",
" <td>34.35</td>\n",
" <td>-8.885942</td>\n",
" <td>8.40</td>\n",
" <td>7.69</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-09-01</th>\n",
" <td>34.65</td>\n",
" <td>0.873362</td>\n",
" <td>7.20</td>\n",
" <td>-14.14</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-12-01</th>\n",
" <td>33.70</td>\n",
" <td>-2.741703</td>\n",
" <td>8.92</td>\n",
" <td>-5.69</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2018-03-01</th>\n",
" <td>37.00</td>\n",
" <td>9.792285</td>\n",
" <td>7.20</td>\n",
" <td>139.53</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2018-06-01</th>\n",
" <td>43.75</td>\n",
" <td>18.243243</td>\n",
" <td>18.60</td>\n",
" <td>121.12</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">9958</th>\n",
" <th>2018-06-01</th>\n",
" <td>54.20</td>\n",
" <td>-14.645669</td>\n",
" <td>0.56</td>\n",
" <td>72.61</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2020-12-01</th>\n",
" <td>128.50</td>\n",
" <td>7.083333</td>\n",
" <td>5.28</td>\n",
" <td>-16.39</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-06-01</th>\n",
" <td>113.50</td>\n",
" <td>-11.673152</td>\n",
" <td>21.08</td>\n",
" <td>43.11</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-09-01</th>\n",
" <td>111.00</td>\n",
" <td>-2.202643</td>\n",
" <td>20.08</td>\n",
" <td>30.52</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-12-01</th>\n",
" <td>112.00</td>\n",
" <td>0.900901</td>\n",
" <td>6.96</td>\n",
" <td>59.19</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>8521 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" 收盤價(元) 報酬% ROE(A)-稅後 營業利益成長率\n",
"證券代碼 年月日 \n",
"1101 2017-06-01 34.35 -8.885942 8.40 7.69\n",
" 2017-09-01 34.65 0.873362 7.20 -14.14\n",
" 2017-12-01 33.70 -2.741703 8.92 -5.69\n",
" 2018-03-01 37.00 9.792285 7.20 139.53\n",
" 2018-06-01 43.75 18.243243 18.60 121.12\n",
"... ... ... ... ...\n",
"9958 2018-06-01 54.20 -14.645669 0.56 72.61\n",
" 2020-12-01 128.50 7.083333 5.28 -16.39\n",
" 2021-06-01 113.50 -11.673152 21.08 43.11\n",
" 2021-09-01 111.00 -2.202643 20.08 30.52\n",
" 2021-12-01 112.00 0.900901 6.96 59.19\n",
"\n",
"[8521 rows x 4 columns]"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data_no"
]
},
{
"cell_type": "code",
"execution_count": 28,
"id": "1c837464",
"metadata": {},
"outputs": [],
"source": [
"import sklearn.preprocessing as preprocessing\n",
"\n",
"data_rmout = data_no.replace([np.inf, -np.inf], np.nan)\n",
"data_rmout = data_no.dropna()\n",
"data_std = pd.DataFrame(preprocessing.scale(data_rmout), index = data_no.index, columns = data_no.columns)"
]
},
{
"cell_type": "code",
"execution_count": 29,
"id": "9aad868d",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th></th>\n",
" <th>收盤價(元)</th>\n",
" <th>報酬%</th>\n",
" <th>ROE(A)-稅後</th>\n",
" <th>營業利益成長率</th>\n",
" </tr>\n",
" <tr>\n",
" <th>證券代碼</th>\n",
" <th>年月日</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">1101</th>\n",
" <th>2017-06-01</th>\n",
" <td>34.35</td>\n",
" <td>-8.885942</td>\n",
" <td>8.40</td>\n",
" <td>7.69</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-09-01</th>\n",
" <td>34.65</td>\n",
" <td>0.873362</td>\n",
" <td>7.20</td>\n",
" <td>-14.14</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2017-12-01</th>\n",
" <td>33.70</td>\n",
" <td>-2.741703</td>\n",
" <td>8.92</td>\n",
" <td>-5.69</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2018-03-01</th>\n",
" <td>37.00</td>\n",
" <td>9.792285</td>\n",
" <td>7.20</td>\n",
" <td>139.53</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2018-06-01</th>\n",
" <td>43.75</td>\n",
" <td>18.243243</td>\n",
" <td>18.60</td>\n",
" <td>121.12</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th rowspan=\"5\" valign=\"top\">9958</th>\n",
" <th>2018-06-01</th>\n",
" <td>54.20</td>\n",
" <td>-14.645669</td>\n",
" <td>0.56</td>\n",
" <td>72.61</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2020-12-01</th>\n",
" <td>128.50</td>\n",
" <td>7.083333</td>\n",
" <td>5.28</td>\n",
" <td>-16.39</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-06-01</th>\n",
" <td>113.50</td>\n",
" <td>-11.673152</td>\n",
" <td>21.08</td>\n",
" <td>43.11</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-09-01</th>\n",
" <td>111.00</td>\n",
" <td>-2.202643</td>\n",
" <td>20.08</td>\n",
" <td>30.52</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2021-12-01</th>\n",
" <td>112.00</td>\n",
" <td>0.900901</td>\n",
" <td>6.96</td>\n",
" <td>59.19</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>8521 rows × 4 columns</p>\n",
"</div>"
],
"text/plain": [
" 收盤價(元) 報酬% ROE(A)-稅後 營業利益成長率\n",
"證券代碼 年月日 \n",
"1101 2017-06-01 34.35 -8.885942 8.40 7.69\n",
" 2017-09-01 34.65 0.873362 7.20 -14.14\n",
" 2017-12-01 33.70 -2.741703 8.92 -5.69\n",
" 2018-03-01 37.00 9.792285 7.20 139.53\n",
" 2018-06-01 43.75 18.243243 18.60 121.12\n",
"... ... ... ... ...\n",
"9958 2018-06-01 54.20 -14.645669 0.56 72.61\n",
" 2020-12-01 128.50 7.083333 5.28 -16.39\n",
" 2021-06-01 113.50 -11.673152 21.08 43.11\n",
" 2021-09-01 111.00 -2.202643 20.08 30.52\n",
" 2021-12-01 112.00 0.900901 6.96 59.19\n",
"\n",
"[8521 rows x 4 columns]"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data_rmout "
]
},
{
"cell_type": "code",
"execution_count": 30,
"id": "3f52af3b",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(array([ 78., 131., 152., 241., 311., 409., 487., 679., 870., 924., 749.,\n",
" 567., 431., 339., 290., 244., 204., 163., 147., 132., 105., 97.,\n",
" 79., 74., 59., 66., 43., 39., 33., 34., 33., 28., 28.,\n",
" 29., 19., 25., 15., 19., 20., 11., 19., 18., 8., 12.,\n",
" 9., 13., 7., 10., 7., 14.]),\n",
" array([-1.55602479, -1.42528892, -1.29455305, -1.16381718, -1.03308131,\n",
" -0.90234544, -0.77160957, -0.6408737 , -0.51013783, -0.37940196,\n",
" -0.24866609, -0.11793022, 0.01280565, 0.14354151, 0.27427738,\n",
" 0.40501325, 0.53574912, 0.66648499, 0.79722086, 0.92795673,\n",
" 1.0586926 , 1.18942847, 1.32016434, 1.45090021, 1.58163608,\n",
" 1.71237195, 1.84310782, 1.97384369, 2.10457956, 2.23531543,\n",
" 2.3660513 , 2.49678717, 2.62752304, 2.75825891, 2.88899478,\n",
" 3.01973065, 3.15046652, 3.28120239, 3.41193826, 3.54267413,\n",
" 3.67341 , 3.80414587, 3.93488174, 4.06561761, 4.19635348,\n",
" 4.32708935, 4.45782522, 4.58856109, 4.71929696, 4.85003283,\n",
" 4.9807687 ]),\n",
" <BarContainer object of 50 artists>)"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 1080x576 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.figure(figsize=(15,8))\n",
"plt.hist(data_std['ROE(A)-稅後'], bins = 50)\n",
"plt.hist(data_std['營業利益成長率'], bins = 50, alpha =0.7)"
]
},
{
"cell_type": "code",
"execution_count": 31,
"id": "7b5c4c4e",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"SVC(C=10, break_ties=False, cache_size=200, class_weight=None, coef0=0.0,\n",
" decision_function_shape='ovr', degree=3, gamma='scale', kernel='linear',\n",
" max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
" tol=0.001, verbose=False)"
]
},
"execution_count": 31,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.svm import SVC\n",
"\n",
"cf = SVC(C=10.0, break_ties=False, cache_size=200, class_weight=None, coef0=0.0,\n",
" decision_function_shape='ovr', degree=3, gamma='scale', kernel='linear',\n",
" max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
" tol=0.001, verbose=False)\n",
"\n",
"index = ['ROE(A)-稅後', '營業利益成長率']\n",
"\n",
"cf.fit(data_train[index], data_train['報酬%'] > data_train['報酬%'].quantile(0.5))"
]
},
{
"cell_type": "code",
"execution_count": 32,
"id": "20791b18",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<AxesSubplot:>"
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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