📊📔 Diario Data Analysis in Transport Systems

📆 15-10-2024 - Tue

📝 Theory

✏️ Projects

📆 22-10-2024 - Tue

📝 Theory

✏️ Projects

📆 24-10-2024 - Thu

LAB Session

📆 29-10-2024 - Tue

📝 Theory

03 - Computational Statistical Inference - DATS

• Contrast of hypothesis - Hypothesis testing
  • Level of significance
• Correlation between 2 numeric variables - $(Y,X)$
  • Charts
  • Numbers
    • Pearson Correlation coefficient
    • Spearman's Correlation Coefficient
  • Test
    • cor.test(X ~ Y) - Person's test
    • cor.test(X ~ Y, method = 'spearman') - Person's test
• Correlation between a numeric variable and a factor (non numeric variable) - $(Y,A)$
  • Charts
    • boxplot(Y ~ A) or just use plot(Y ~ A)
  • Numbers
    • Coefficient of Determination $R^2⟵$ lm(Y~A) (Linear Model)
    • $et{a}^2$
  • Test
    • Test on means
      • j = 2
        • Parametric - Welch test: t.test(Y~A)
        • Non-parametric - Wilcoxon Test: wilcox.test(Y~A)
      • j > 2
        • Parametric - oneway.test(Y~A)
        • Non-parametric - kruskal.test(Y~A)
    • Test on variance
      • j = 2
        • Parametric - var.test(Y~A)
        • Non-parametric - fligner.test(Y~A)
      • j > 2
        • Parametric - barlett.test(Y~A)
        • Non-parametric - fligner.test(Y~A)
• Correlation between 2 factors - $(B,A)$
  • Charts
    • mosaic plot - plot(B~A) You get mosaic plot automatically if it's 2 factos
  • Numbers
    • Contingency tables - table(A,B) --> prob.table((A,B), 1) or prob.table((A,B), 2)
  • Test
    • Pearson's test on independence or ${\chi }^2$ test - chisq.test(B~A)

📆 31-10-2024 - Thu

✏️ Lab session

LabSess 3 - 2024-10-31 - Computational Statistical Inference - DATS

We practiced all the tools introduced in the class of 2024-10-29 into RStudio.

📆 05-11-2024 - Tue

📝 Theory

03 - Computational Statistical Inference - DATS

• pariwise.t.test(Y,A) - pariwise.wilcox.test(Y~A)
• library(FactoMineR)
  • condes(df,'nb') and catdes(df,'nb')
• ExTheo_5nov2024.Rmd example

📆 12-11-2024 - Tue

Preparation for Midterm.
Work in class on Midterm 2023-2024.

📆 14-11-2024 - Thu

• Preparation for Midterm.
  • Work in class on Midterm 2023-2024.
• Kyoto data analysis

📆 19-11-2024 - Tue

Midterm exam!

📆 26-11-2024 - Tue

📝 Theory

04 - Statistical modeling through regression - DATS

• Linear models
  • Least square method

📆 28-11-2024 - Thu

📝 Theory

Lab session

📆 03-12-2024 - Tue

📝 Theory

04 - Statistical modeling through regression - DATS

• Interpretation of lm() method output in R-Studio.
• Fisher test
  • 3 ways of defining the fisher test
• Testing multiple variables at the same time: anova(submodel,Model)
  • Output interpretation
  • Multiple correlation coefficient
• Prediction based on linear model
  • predict(<model>, newdata=<dataframe>, interval = "prediction")
• Model validation
  • plot(<model>) interpretation

📆 10-12-2024 - Tue

📝 Theory

04 - Statistical modeling through regression - DATS

• Residuals
• Diagnostics
• Picking Best model

📆 12-12-2024 - Thu

Lab session on multiple linear regression.

📆 17-12-2024 - Tue

📝 Theory

❗❗❗❗❗❗❗❗❗❗❗❗
❗❗❗ COMPLETARE ❗❗❗
❗❗❗❗❗❗❗❗❗❗❗❗