Using the new historical election data, I construct a 3-year rolling-window own-party jacknife-vote-share feature in the election model and we see adjusted R squared increase from 0.067 to 0.098.
This markdown contains table 01s for elections with a new Election LM model. Here I make use of the new historical data from Rowan, which allows me to construct a rolling window feature for own party vote share. I am making use of a three year window looking back, and excluding the current election year. This is constructed at the house of representatives level as these have the most years.
NOTE In this table, to allow for as much data as possible, I am using the older 4-voter side-by-side MTurk features without quality control.
Before jumping into regression output, I will put all definitions of important terms here:
Vote-Share : In all tables below, when vote-share is a LHS variable this means the vote-share of a specific candidate election. I.e. how much of the total vote did a specific candidate get in their election.
Own-Party-Vote-Share-Jacknife : This is a feature and computed as the “average vote share of my party in this states house of rep. election over the previous 3 years excluding the current”
Replacing own_party_win_rate with own_party_vote_share_jacknife in the election lm. This now includes:
NOTE We see a significant increase in the coefficient on the election_lm from 0.067 to 0.098. Previous election models used the vote share over the entire period, and did not have a rolling window function.
Own Party Vote-share-Jacknife is computed as: “average vote share of my party in this states house of rep. election over the previous 3 years excluding the current”
| Table 01 - Version 04 - Election Regressions | |||
|---|---|---|---|
| Fit measured in adjusted R squared and AUC | |||
| Model Configuration | Election Outcome | Vote Share | |
| Adjusted R Squared | ROC AUC | Adjusted R squared | |
| Single Variable Model | |||
| Election LM | 0.0977 | 0.6803 | 0.0877 |
| Lower 95% C.I. | 0.0716 | 0.6496 | 0.0628 |
| Upper 95% C.I. | 0.1275 | 0.7111 | 0.1152 |
| Sex | 0.0034 | 0.5150 | 0.0047 |
| 0.0003 | 0.4916 | 0.0001 | |
| 0.0096 | 0.5383 | 0.0130 | |
| Skine-Tone | −0.0086 | 0.5422 | −0.0044 |
| −0.0035 | 0.5092 | −0.0006 | |
| 0.0187 | 0.5752 | 0.0287 | |
| MTurk Features | 0.0154 | 0.5653 | 0.0008 |
| 0.0073 | 0.5321 | −0.0017 | |
| 0.0332 | 0.5986 | 0.0128 | |
| P_hat_cnn | 0.0988 | 0.6837 | 0.0770 |
| 0.0731 | 0.6529 | 0.0527 | |
| 0.1275 | 0.7144 | 0.1046 | |
| Combined Variable Model | |||
| Election LM + P_hat_cnn | 0.1883 | 0.7546 | 0.1598 |
| 0.1575 | 0.7267 | 0.1307 | |
| 0.2222 | 0.7824 | 0.1925 | |
| Election LM + Sex | 0.1001 | 0.6835 | 0.0923 |
| 0.0750 | 0.6529 | 0.0694 | |
| 0.1288 | 0.7142 | 0.1225 | |
| Election LM + Sex + P_hat_cnn | 0.1900 | 0.7563 | 0.1634 |
| 0.1587 | 0.7286 | 0.1357 | |
| 0.2265 | 0.7841 | 0.1986 | |
| Election LM + Sex + Skin-Tone | 0.0920 | 0.6907 | 0.0869 |
| 0.0809 | 0.6603 | 0.0779 | |
| 0.1399 | 0.7212 | 0.1310 | |
| Election LM + Sex + Skin-Tone + P_hat_cnn | 0.1832 | 0.7604 | 0.1593 |
| 0.1650 | 0.7329 | 0.1432 | |
| 0.2335 | 0.7879 | 0.2104 | |
| Election LM + Sex + Skin-Tone + MTurk | 0.1036 | 0.7048 | 0.0856 |
| 0.0932 | 0.6749 | 0.0767 | |
| 0.1540 | 0.7348 | 0.1369 | |
| Election LM + Sex + Skin-Tone + MTurk + P_hat_cnn | 0.1866 | 0.7656 | 0.1568 |
| 0.1678 | 0.7382 | 0.1442 | |
| 0.2377 | 0.7929 | 0.2118 | |