archiewood

select
  date,
  contractname as candidate,
  substring(closeshareprice,2,10)::double as implied_probability
from read_csv('https://www.predictit.org/Resource/DownloadMarketChartData?marketid=7456')
order by date desc, implied_probability desc
limit 8

archiewood

with cleaned_docs as (
    select 
        tweet_id,
        regexp_replace(regexp_replace(text, '\\n', ' ', 'g'), 'https[^\\s]+', '', 'g') as cleaned_text
    from tweets
    where text is not null
),

tokenized_docs as (
    select 

archiewood

INSTALL 'icu';
LOAD 'icu';

SET TimeZone = 'Europe/Berlin';


create or replace table metrics as SELECT TIMESTAMP '2024-06-07 00:00:00' AS mytimestamp, TIMESTAMP WITH TIME ZONE '2024-06-07 00:00:00+02' AS mytimestamptz, 1000 AS sales, 50 AS orders, 20 AS customers;

-- Check the stored timestamps to ensure they match Berlin time
SELECT 

archiewood

SELECT
    regr_slope(num_orders, sales) AS slope,
    regr_intercept(num_orders, sales) AS intercept,
    regr_r2(num_orders, sales) AS r_squared
FROM orders_by_state

archiewood

/** 
 * Reads the manifest file and creates a schema and table for each parquet file in the manifest.
 * It will create a DB called `local.duckdb` in the root of the project.
 *  
 * Usage:
 * 1. Copy this file to the root of your project Evidence project.
 * 2. `npm run sources`
 * 3. `node createLocalDuckDB.js`
 */