Logistic-regression-in-python

01_LR_Introduction (Theory)

• Predict categories based on MLE

02_Odd_LogOdd_OddRatio (Theory)

• Probability : Something Happening / Everything that could Happen
• Odds : Something Happening / Something Not Happening
• Log(Odds) : To make Odds output symmetry

03_Logit_Model (Theory)

• Indepth Logistic Output explained

04_Likelihood_Probability (Theory)

05_MLE (Theory)

06_LR_Assumptions (Theory)

• Assumption 1 - Appropriate outcome type (Must be categorical)
• Assumption 2 - Linearity of independent variables and log odds
• Assumption 3 - No strongly influential outliers
• Assumption 4 - Absence of multicollinearity
• Assumption 5 - Independence of observations
• Assumption 6 - Sufficiently large sample size

07_LR_Assumptions (Python Code)

• Python Code for Logistic Regression Assumptions

08_AIC_BIC (Theory)

• Akaike Information Criterion
• Bayesian Information Criterion
• Choose the lowest score

09_Logistic_Regression (Python Code)

• Python Code for Logistic Regression

10_Multiclass_Classification (Theory)

• One vs All (OvA) also known as One vs Rest (OvR)
• One vs One (OnO)

11_Multi_Class_Classification (Python Code)

• Python Code for Multi Class Classification

12_Regularization (Theory)

• L1 Lasso
  • SSR + lamda * (slope)^2
  • Useless variable become 0
• L2 Ridge
  • SSR + lamda * |slope|
  • Useless variable tends to become 0 but never = 0
• Elastic Net : Combination of L1 & L2

13_LR_Regularization (Python Code)

• Python Code of Regularization (L1 Lasso,L2 Ridge & Elastic Net)

14_WOE_IV (Theory)

• Weight of Evidence : Predictive power of Independent Variables
• Information Value : Technique to select important Variables

15_LR_WOE_IV (Python Code)

• Python Code for WOE and IV

16_LR_Revision (Theory)

• Logistic Regression Revision

17_LR_1_Interview_Questions (Theory)

• Logistic Regression Interview quesion bank

18_LR_2_Interview_Questions (Theory)

• Indepth Logistic Output explained