Readings

[twang15]

The Actual Difference Between Statistics and Machine Learning

[twang15]

Machine Learning vs. Deep Learning

Both Machine Learning and Deep Learning are able to handle massive dataset sizes, however, machine learning methods make much more sense with small datasets.

[twang15]

one-hot encoding vs. dummy encoding in scikit-learn

[twang15]

What is one-hot encoding

[twang15]

Multiple linear regression w/ one categorical variable

[twang15]

Multiple linear regression w/ two dummy variables 1 Multiple linear regression w/ two dummy variables 2

[twang15]

One-Hot-Encoding, Multicollinearity and the Dummy Variable Trap

• multicollinearity is generally measured using Variance Inflation Factor

[twang15]

VIF: Measuring the degree of multicollinearity

• The squared multiple-correlation coefficient for a specific parameter is a measure of the linear strength between a variable X and the rest of the variables included in the model.

[twang15]

SPSS Chi-square and Fisher test

SPSS student-t test w/ or w/o homogeneous variance

SPSS ANOVA test and Levene Test for homogeneous variance test

[twang15]

超详细SPSS操作：多重线性回归（完整版）

[twang15]

Calculating Unstandardized and Standardized Predicted and Residual Values in SPSS and Excel

Deleted residual, Studentized residual, and outlier detection

[twang15]

Diagnose multi-collinearity Diagnose multi-collinearity 2

[twang15]

Overfitting

• be sure to state carefully what assumptions you made, what decisions you made, your basis for making these decisions, and what validation procedures you used. Provide (in supplementary online material if necessary) enough detail so that another researcher could replicate your methods.

[twang15]

6种统计学方法，让你666地发文章

[twang15]

Detecting Multicollinearity Using Variance Inflation Factors

• The VIF for the predictor Weight, for example, tells us that the variance of the estimated coefficient of Weight is inflated by a factor of 8.42 because Weight is highly correlated with at least one of the other predictors in the model
• If the variances of the estimated coefficients are inflated by multicollinearity, then our confidence intervals for the slope parameters are wider and therefore less useful. Eliminating or even reducing the multicollinearity therefore yields narrower, more useful confidence intervals for the slopes.
• The main reason for centering to correct structural multicollinearity is that low levels of multicollinearity can be helpful in avoiding computational inaccuracies. Specifically, a near-zero determinant of XTX is a potential source of serious roundoff errors in the calculations of the normal equations. Severe multicollinearity has the effect of making this determinant come close to zero. Thus, under severe multicollinearity, the regression coefficients may be subject to large roundoff errors.

[twang15]

Modeling Binary logistic Regression Using Python Modeling Binary logistic Regression Using R