Package com.leumanuel.woozydata.service

Class StatisticalService

java.lang.Object
com.leumanuel.woozydata.service.StatisticalService
public class StatisticalService extends Object
Service class for advanced statistical operations and hypothesis testing. Provides implementation of various statistical tests and data analysis methods.
Version:
1.0
Author:
Leu A. Manuel

  • Constructor Details

    • StatisticalService

      public StatisticalService()

  • Method Details

    • tTest

      public Map<String,Double> tTest(DataFrame df, String col1, String col2)
      Performs t-test between two columns.
      Parameters:
      df - DataFrame containing the data
      col1 - First column name
      col2 - Second column name
      Returns:
      Map containing test results

    • anova

      public Map<String,Double> anova(DataFrame df, String... columns)
      Performs ANOVA test on multiple columns.
      Parameters:
      df - DataFrame containing the data
      columns - Column names to test
      Returns:
      ANOVA test results

    • chiSquareTest

      public Map<String,Double> chiSquareTest(DataFrame df, String col1, String col2)
      Performs chi-square test for independence.
      Parameters:
      df - DataFrame containing the data
      col1 - First categorical column
      col2 - Second categorical column
      Returns:
      Chi-square test results

    • shapiroWilkTest

      public Map<String,Double> shapiroWilkTest(DataFrame df, String column)
      Performs Shapiro-Wilk test for normality on a numeric column. Tests the null hypothesis that the data is normally distributed.
      Parameters:
      df - DataFrame containing the data
      column - Column name to test for normality
      Returns:
      Map containing 'statistic' (W) and 'p_value'
      Throws:
      IllegalArgumentException - if column is not numeric

    • select

      public DataFrame select(DataFrame df, String... columns)
      Selects specified columns from the DataFrame. Creates a new DataFrame containing only the selected columns.
      Parameters:
      df - Source DataFrame
      columns - Columns to select
      Returns:
      New DataFrame with only selected columns
      Throws:
      IllegalArgumentException - if any column doesn't exist

    • sample

      public DataFrame sample(DataFrame df, int n)
      Creates a random sample of specified size from the DataFrame. Uses random sampling without replacement.
      Parameters:
      df - Source DataFrame
      n - Sample size
      Returns:
      New DataFrame containing the random sample
      Throws:
      IllegalArgumentException - if n is larger than DataFrame size