Instructions: Upload a CSV file containing your historical price data. The CSV should have at least two columns: 'time' (for timestamp/date) and 'value' (for price). The script will calculate log returns, descriptive statistics, and autocorrelation functions for both returns and squared returns.
Analysis Results
Data Summary
Number of Data Points:
Time Period:
Log Returns Descriptive Statistics
| Statistic | Log Returns | Raw Prices |
|---|---|---|
| Mean | ||
| Standard Deviation (Volatility) | ||
| Skewness | ||
| Kurtosis | ||
| Last Price Value | ||
*Kurtosis values significantly greater than 3 indicate "fat tails" (more extreme events than a normal distribution).
Autocorrelation Function (ACF) of Log Returns
Values close to zero for all lags suggest a random walk. Significant values indicate predictability.
Autocorrelation Function (ACF) of Squared Log Returns (for Volatility Clustering)
Significant values indicate volatility clustering (conditional heteroskedasticity).
Monthly Volatility (Standard Deviation of Log Returns)
This shows how volatility changes month-to-month.
Monthly Volatility (Standard Deviation of Raw Prices)
Automated Recommendation
Generating recommendations based on computed statistics...
Error: