udpate

2020MAR30/notes.md

@@ -14,9 +14,10 @@
 ### 2) using residual basis of the self subject and the coefficients of other subjects to make predictions
 ##### Using residual bases to make prediction differentiates self from others more than using original basis
  * left panel: prediction using original basis; right panel: prediction using residual basis
+
 <img src="figs/improv.png" width="49%"> <img src="figs/improv_reg.png" width="49%"> 
  * but if add group effects back on:
-<img src="figs/improv.png" width="49%"> <img src="figs/improv_reg.png" width="49%">
+<img src="figs/improv.png" width="49%"> <img src="figs/improv_reg_addback.png" width="49%">
  * so it is probably biased by group level effects...
 ##### using residual bases to make prediction also improves prediction accuracy
 ![](../2020MAR24/figs/model_accuracy.png)