Advanced Controls

Probably offers a range of advanced controls that allow you to customize your analysis, fine-tune your visualizations, and dive deeper into your data. This guide will walk you through these advanced features and how to use them effectively.

Statistical Controls

Significance Level Adjustment

1. Open the "Advanced Controls" panel.
2. Locate the "Significance Level" slider.
3. Adjust the slider to set your desired p-value threshold (default is 0.05).
4. This affects which results are considered statistically significant in your analysis.

Effect Size Calculation

1. In the "Advanced Controls" panel, find the "Effect Size" section.
2. Choose between Cohen's d, Hedge's g, or Glass's Δ for continuous variables.
3. For categorical variables, select from Cramér's V, Phi coefficient, or odds ratio.
4. The chosen effect size measure will be displayed alongside your statistical test results.

Visualization Customization

Color Scheme Selection

1. Open the "Plot Settings" in the Advanced Controls panel.
2. Click on "Color Scheme" to open the color palette options.
3. Choose from pre-defined color schemes or create a custom palette.
4. Apply different color schemes to distinguish between variables or highlight specific data points.

Axis Scaling

1. In the "Plot Settings", find the "Axis Scaling" options.
2. Choose between linear, logarithmic, or custom scaling for each axis.
3. For custom scaling, you can input a mathematical function to transform your data.

Plot Type Conversion

1. Use the "Plot Type" dropdown in the main interface to switch between different visualization types.
2. Some advanced options include:
   • Violin plots for distribution visualization
   • Sunburst charts for hierarchical data
   • Sankey diagrams for flow visualization

Data Transformation

Custom Calculations

1. Open the "Data Transformation" tab in Advanced Controls.
2. Click "Add Calculation" to create a new derived variable.
3. Use the formula editor to write custom calculations using existing variables.
4. Your new calculated variable will be available for plotting and analysis.

Binning and Discretization

1. In "Data Transformation", select a continuous variable.
2. Choose "Bin" to convert it into a categorical variable.
3. Select from methods like equal width, equal frequency, or custom binning.
4. Adjust the number of bins or bin edges as needed.

Advanced Clustering

Algorithm Selection

1. When in Cluster View, open the "Clustering Settings" in Advanced Controls.
2. Choose from algorithms like K-means, DBSCAN, or Hierarchical Clustering.
3. Adjust algorithm-specific parameters (e.g., number of clusters for K-means).

Dimensionality Reduction

1. In "Clustering Settings", find the "Dimensionality Reduction" options.
2. Choose between techniques like PCA, t-SNE, or UMAP.
3. Adjust parameters like perplexity for t-SNE or n_neighbors for UMAP.
4. This affects how high-dimensional data is projected into 2D or 3D for visualization.

Time Series Analysis

Seasonality Decomposition

1. For time series data, open the "Time Series" tab in Advanced Controls.
2. Select "Seasonality Decomposition" to break down your time series into trend, seasonal, and residual components.
3. Choose between additive or multiplicative decomposition models.

Moving Averages

1. In the "Time Series" tab, select "Moving Average".
2. Choose the window size for your moving average.
3. Select between simple moving average (SMA) or exponential moving average (EMA).

Export and Reproducibility

Code Generation

1. After performing your analysis, click on "Generate Code" in the Advanced Controls panel.
2. Choose your preferred language (Python, R, or SQL).
3. Probably will generate code that reproduces your current analysis and visualizations.

Custom Reporting

1. In the "Export" section of Advanced Controls, click on "Create Report".
2. Select which plots and statistical results to include in your report.
3. Choose between PDF, HTML, or Jupyter Notebook formats for your report.

By mastering these advanced controls, you can take your data analysis to the next level, customizing every aspect of your exploration and ensuring that your visualizations and insights are precisely tailored to your needs. Remember, with great power comes great responsibility – always ensure that your advanced analyses are statistically sound and interpretable within the context of your data and research questions.