Linear Regression Ridge

Ridge Regression Report

Cases Used in the Analysis

Model: y ~ age + sex + bmi + map + tc + ldl + hdl + tch + ltg + glu

Data Used No.
Number of cases read 442
Number of cases used in the analysis 442
Number of incomplete (omitted) cases 0

Ridge Coefficient Shrinkage Paths Across Lambda Values

Model: y ~ age + sex + bmi + map + tc + ldl + hdl + tch + ltg + glu

A graph of Ridge Coefficient Shrinkage Paths Across Lambda Values

K-Fold Cross-Validation: Mean RMSE vs. Lambda Values

Model: y ~ age + sex + bmi + map + tc + ldl + hdl + tch + ltg + glu
Number of Folds = 10, Seed = 100
Optimal Lambda = 17.624
95% Confidence Interval for RMSE at Optimal Lambda = (49.871, 59.165)
One Standard Error Lambda Threshold = 45.109

A graph of Mean R M S Ein k-fold cross validation Versus Lambda and Log of Lambda. The optimal value of lambda is 17.6237821787457

Parameter Estimates at the RMSE-Optimal Lambda (Original Data Scale)

Model: y ~ age + sex + bmi + map + tc + ldl + hdl + tch + ltg + glu
Estimates at Lambda = 17.623782

Parameter Estimate Parameter Estimate
(Intercept) 152.13348 glu 77.43721
age -3.61315
sex -224.34086
bmi 511.21687
map 313.56044
tc -161.63799
ldl -19.816
hdl -166.64326
tch 113.95283
ltg 496.27731

Parameter Estimates at the RMSE-Optimal Lambda (Standardized Data Scale)

Model: y ~ age + sex + bmi + map + tc + ldl + hdl + tch + ltg + glu
Estimates at Lambda = 17.623782

Parameter Estimate
age -1.66263
sex -9.53072
bmi 24.65258
map 15.76091
tc -7.90786
ldl 0.22299
hdl -7.47723
tch 4.60569
ltg 24.85567
glu 3.92681