regression - How to perform a 70/30 holdout in R -


i'm trying make prediction based on below linear regression model:

enter image description here

i want predict intercept (market share) based on formula given value each of response variables. formula these results , can plug numbers in each variable?

or need holdout/training sets first?

edit: added text of results.

summary(fit2)  call: lm(formula = headache_panel_cleaned$private_label_cleaned ~ income_cleaned +      age_cleaned + education_cleaned, data = headache_panel_cleaned)  residuals:     min      1q  median      3q     max  -53.880 -33.804  -5.473  32.589  68.171   coefficients:                   estimate std. error t value pr(>|t|)     (intercept)       52.02867    0.96849  53.721   <2e-16 *** income_cleaned    -0.22711    0.01199 -18.949   <2e-16 *** age_cleaned       -0.11363    0.01334  -8.516   <2e-16 *** education_cleaned  0.13104    0.01213  10.807   <2e-16 *** --- signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1  residual standard error: 35.09 on 38115 degrees of freedom   (8224 observations deleted due missingness) multiple r-squared:  0.0102,    adjusted r-squared:  0.01012  f-statistic:   131 on 3 , 38115 df,  p-value: < 2.2e-16 


Comments

Popular posts from this blog

angular - Ionic slides - dynamically add slides before and after -

minify - Minimizing css files -

Add a dynamic header in angular 2 http provider -