# The Math Behind RVI ## Components of RVI The **Relative Volatility Index (RVI)** measures the direction of volatility in the market, using components like: - Standard deviation of price changes - Simple moving average (SMA) to smooth volatility - Separation of up and down price movements ### RVI Formula The RVI is calculated using the following formula: $$ \text{RVI}_t = 100 \times \frac{\text{SMA}(\sigma_{\text{up}}, N)}{\text{SMA}(\sigma_{\text{up}}, N) + \text{SMA}(\sigma_{\text{down}}, N)} $$ Where: - \( \text{RVI}_t \) is the RVI value at time \( t \) - \( \sigma_{\text{up}} \) is the standard deviation of up moves over the lookback period \( N \) - \( \sigma_{\text{down}} \) is the standard deviation of down moves over the lookback period \( N \) - \( \text{SMA} \) represents the simple moving average applied over \( N \) periods ### Up and Down Move Calculation The standard deviations \( \sigma_{\text{up}} \) and \( \sigma_{\text{down}} \) are calculated based on the price changes: $$ \Delta \text{Price} = \text{Close}_t - \text{Close}_{t-1} $$ - If \( \Delta \text{Price} > 0 \), it contributes to \( \sigma_{\text{up}} \) - If \( \Delta \text{Price} < 0 \), it contributes to \( \sigma_{\text{down}} \) ### Parameter Definitions RVI uses the following main parameters: - **Lookback period** (\( N \)): The number of periods used to calculate the standard deviations and SMAs. A typical value is 14. - **Smoothing with SMA**: The standard deviations of up and down moves are smoothed using a simple moving average (SMA), making the RVI less sensitive to short-term fluctuations. ### Computational Process For each new data point: - Calculate the price change (\( \Delta \text{Price} \)) from the previous period. - Separate the price changes into up moves and down moves. - Compute the standard deviations (\( \sigma_{\text{up}} \) and \( \sigma_{\text{down}} \)) over the last \( N \) periods. - Apply the simple moving average (SMA) to both up and down standard deviations. - Use the RVI formula to produce the final RVI value.