Learn R Programming

fBasics (version 4041.97)

BasicStatistics: Basic time series statistics

Description

Computes basic financial time series statistics.

Usage

basicStats(x, ci = 0.95)

Value

a data frame with one column for each column of x and the following rows:

"nobs"

number of observations,

"NAs"

number of NAs

"Minimum"

minimum,

"Maximum "

maximum,

"1. Quartile"

lower quartile,

"3. Quartile"

upper quartile,

"Mean"

mean,

"Median"

median,

"Sum"

sum of the values,

"SE Mean"

standard error of the mean,

"LCL Mean"

lower limit of the CI for the mean,

"UCL Mean"

upper limit of the CI for the mean,

"Variance"

variance,

"Stdev"

standard deviation,

"Skewness"

skewness coefficient,

"Kurtosis"

excess kurtosis.

Arguments

x

an object of class "timeSeries" or any other object which can be transformed by the function as.timeSeries into an object of class "timeSeries". The latter case, other than "timeSeries" objects, is more or less untested.

ci

confidence interval, a numeric value, by default 0.95, i.e. 95%.

Details

Computes a number of sample statistics for each column of x. The statistics should be clear from the row names of the returned data frame.

"LCL" and "UCL" stand for lower/upper confidence limits, computed under the null hypothesis of i.i.d.

"Kurtosis" represents the excess kurtosis, so its theoretical value for the normal distribution is zero, not 3.

These statistics are often computed as a first step in the study of returns on financial assets. In that case any inference on these statistics (including the confidence intervals for the mean) should be considered exploratory, since returns are virtually never i.i.d.

Examples

Run this code
set.seed(1234)
## Simulated Monthly Return Data
tS <- timeSeries(matrix(rnorm(12)), timeDate::timeCalendar())
basicStats(tS)

Run the code above in your browser using DataLab