Confidence vs. Unsigned PEs

First prediction confidence vs. Unsigned PEs

## 
## Call:
## lm(formula = conf_1 ~ `Unsigned PE`, data = df.1)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -61.681 -12.595   3.231  18.133  42.076 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   62.13199    0.89351  69.537   <2e-16 ***
## `Unsigned PE` -0.08416    0.05971  -1.409    0.159    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 23.94 on 1682 degrees of freedom
##   (436 observations deleted due to missingness)
## Multiple R-squared:  0.00118,    Adjusted R-squared:  0.0005859 
## F-statistic: 1.987 on 1 and 1682 DF,  p-value: 0.1589

Second prediction confidence vs. Unsigned PEs

## 
## Call:
## lm(formula = conf_2 ~ `Unsigned PE`, data = df.2)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -55.417 -13.315  -1.122  19.231  45.663 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   55.48230    1.12849  49.165   <2e-16 ***
## `Unsigned PE` -0.01637    0.07293  -0.224    0.822    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 24.7 on 1129 degrees of freedom
##   (1021 observations deleted due to missingness)
## Multiple R-squared:  4.461e-05,  Adjusted R-squared:  -0.0008411 
## F-statistic: 0.05037 on 1 and 1129 DF,  p-value: 0.8225