Module saePisan.service.exploration.NormalityTest

Functions

def run_normality_test(parent)

Expand source code

def run_normality_test(parent):
    """
    Runs a normality test on the combined data from two models in the parent object.
    This function performs the following steps:
    1. Activates the R environment.
    2. Retrieves data from two models in the parent object and concatenates them horizontally.
    3. Drops any rows with null values.
    4. Converts the concatenated data to an R dataframe.
    5. Executes an R script provided by the parent object to perform normality tests.
    6. Collects the results of the normality tests and any generated plots.
    Parameters:
    parent (object): An object that contains the following attributes:
        - activate_R(): A method to activate the R environment.
        - model1: An object with a get_data() method that returns a dataframe.
        - model2: An object with a get_data() method that returns a dataframe.
        - r_script (str): An R script to be executed.
        - selected_columns (list): A list of column names to be tested for normality.
        - result (str): A string to store the results of the normality tests.
        - plot (list): A list to store the paths of generated plots.
        - error (bool): A boolean to indicate if an error occurred.
    Raises:
    Exception: If any error occurs during the execution of the R script or data processing.
    """
    
    parent.activate_R()
    df1 = parent.model1.get_data()
    df2 = parent.model2.get_data()
    df = pl.concat([df1, df2], how="horizontal")
    df = df.filter(~pl.all_horizontal(pl.all().is_null()))
    get_data(parent,df)

    try:
        ro.r('rm(list=ls()[ls() != "r_df"])')
        ro.r('suppressMessages(library(nortest))')
        ro.r('suppressMessages(library(tseries))')
        ro.r('suppressMessages(library(ggplot2))')

        ro.r('data <- as.data.frame(r_df)')

        script = parent.r_script
        ro.r(script)

        selected_vars = parent.selected_columns
        plot_paths = []  
        test_names = ["shapiro", "jarque", "lilliefors"]

        # Prepare data containers
        shapiro_data = []
        jarque_data = []
        lillie_data = []

        test_map = {
            "shapiro": shapiro_data,
            "jarque": jarque_data,
            "lilliefors": lillie_data
        }

        for var in selected_vars:

            safe_var = var.replace(" ", "_")

            for test in test_names:
                result_key = f"normality_results_{safe_var}_{test}"
                if ro.r(f"exists('{result_key}')")[0]:
                    stat = ro.r(f"{result_key}$statistic")[0]
                    pval = ro.r(f"{result_key}$p.value")[0]
                    test_map[test].append({
                        "Variable Name": var,
                        "Statistic": stat,
                        "P-Value": pval
                    })

            # Save plots if available
            for plot_type in ["histogram", "qqplot"]:
                plot_name = f"{plot_type}_{safe_var}"
                if ro.r(f"exists('{plot_name}')")[0]:
                    plot_path = f"temp_{plot_name}.png"
                    grdevices.png(file=plot_path, width=800, height=600)
                    ro.r(f"print({plot_name})")
                    grdevices.dev_off()
                    plot_paths.append(plot_path)

        # Build parent.result only with non-empty tables
        result_dict = {}
        if shapiro_data:
            result_dict["Shapiro Test Results"] = pl.DataFrame(shapiro_data)
        if jarque_data:
            result_dict["Jarque Test Results"] = pl.DataFrame(jarque_data)
        if lillie_data:
            result_dict["Lilliefors Test Results"] = pl.DataFrame(lillie_data)

        parent.result = result_dict
        parent.plot = plot_paths if plot_paths else None

        
    except Exception as e:
        parent.result = str(e)
        parent.error = True

Runs a normality test on the combined data from two models in the parent object. This function performs the following steps: 1. Activates the R environment. 2. Retrieves data from two models in the parent object and concatenates them horizontally. 3. Drops any rows with null values. 4. Converts the concatenated data to an R dataframe. 5. Executes an R script provided by the parent object to perform normality tests. 6. Collects the results of the normality tests and any generated plots. Parameters: parent (object): An object that contains the following attributes: - activate_R(): A method to activate the R environment. - model1: An object with a get_data() method that returns a dataframe. - model2: An object with a get_data() method that returns a dataframe. - r_script (str): An R script to be executed. - selected_columns (list): A list of column names to be tested for normality. - result (str): A string to store the results of the normality tests. - plot (list): A list to store the paths of generated plots. - error (bool): A boolean to indicate if an error occurred. Raises: Exception: If any error occurs during the execution of the R script or data processing.