gistfile1.txt

#create a DF to hold errors
search_df = pd.DataFrame()

#set up the grid (1 row x 5 columns)

fig, axs = plt.subplots(1, 5, sharey=True, gridspec_kw={'wspace': 0})

fig.set_size_inches(16,6)

x = y = 0

for issue in myCountries:
  train_l = len(time_series)-5

  selected_series = time_series[[col for col in time_series.columns if (col.find(issue) > -1)]]


  s_model = SARIMAX(endog = selected_series[[issue]][:train_l],
                    exog = selected_series[[x for x in selected_series.columns if x != issue]][:train_l],
                    order=(3,1,1), seasonal_order=(1,0,1,7)).fit()


  f_ru = selected_series[[issue]].copy()[1:]
  f_ru.columns = ["actual"]
  f_ru["predicted"] = s_model.predict(end=datetime.datetime(2019, 10, 6), endog = selected_series[[issue]][-5:],exog = selected_series[[x for x in selected_series.columns if x != issue]][-5:], 
                                            dynamic= False)
  
  testing = f_ru.copy()

  testing["error"] = np.abs((testing["actual"] - testing["predicted"]) / testing["actual"])
  fit = round(testing[testing["actual"] != 0].error.mean()*100)
  testing2 = testing[-5:]
  fit_p = round(testing2[testing2["actual"] != 0].error.mean()*100)

  search_df.loc[issue, "topic_only_model"] = fit
  search_df.loc[issue, "topic_only_predicted"] = fit_p



  f_ru["actual"].plot(title="{}\nMAPE: test: {}% model: {}%".format(issue, fit_p, fit), ax=axs[x])
  f_ru["predicted"][:-5].plot(color="orange", label="predicted: Train", ax=axs[x])
  f_ru["predicted"][-6:].plot(color="red", label="predicted: Test", ax=axs[x])

  selected_series[[x for x in selected_series.columns if x != issue]].plot(style=":",ax=axs[x])

  if x ==0:
    handles, labels = axs[0].get_legend_handles_labels()
    for i in range(len(labels)):
      if labels[i].find("_") > -1:
        labels[i] = labels[i][:labels[i].find("_")]+"     "

  axs[x].get_legend().remove()



  x+=1

fig.tight_layout()

lgd = fig.legend(handles, labels, loc='center right', bbox_to_anchor=(1.15,.5))



axs[0].set_ylabel("% Search or # Articles")
fig.suptitle("Model and Predicted Google Search Results", y=1.05)