Description Usage Arguments Details Value
Extract realized probability distribution from the time series data of the underlying price.
1 2 | realizedDistribution(x, t, n = 512, bounds = c(-0.2, 0.2),
normalization = TRUE)
|
x |
Time series of underlying prices in xts format. The frequency is assumed to be daily. |
t |
Maturity (in years). This is used to scale the daily returns using a square root of time scaling. |
n |
Number of division for the range of strikes to generate the distribution. This number should ideally be a power of 2 (becuase of Fourier transform used within) although it is not a requirement. |
bounds |
The range (in percentage) of the underlying to estimate the distribution. |
normalization |
A logical value. If true, the implied distribution is normalized to sum up to 1, as well as the first moment is normalized to be equal to the forward level (100). |
Extracting probability density from the prices data is done in two steps. First the prices are converted to returns and appropriately scaled. Then these scaled returns are use to fit a (gaussian) kernel density estimate. The probability distribution is normalized in the zero-th moment to sum up to 1 and also in the first moment to match the forward (100). The strikes of the underlying of the distribution is always scaled so that the current forward is at 100.
A dataframe with strikes and corresponding probability density function estimates.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.