Obtaining statistics

The methods listed below are designed to obtain descriptive statistics for matrices and vectors. All of them apply to a vector or a matrix as a whole, as well as to a given matrix axis (horizontally or vertically). When applied entirely to an object, these functions return a scalar (singular). When applied to a matrix along any of the axes, a vector is returned.

The general appearance of prototypes:

T vector<T>::Method(const vector<T> &v)

T matrix<T>::Method(const matrix<T> &m)

vector<T> matrix<T>::Method(const matrix<T> &m, const int axis)

The list of methods:

• ArgMax, ArgMin: find indexes of maximum and minimum values
• Max, Min: find the maximum and minimum values
• Ptp: find a range of values
• Sum, Prod: calculate the sum or product of elements
• CumSum, CumProd: calculate the cumulative sum or product of elements
• Median, Mean, Average: calculate the median, arithmetic mean, or weighted arithmetic mean
• Std, Var: calculate standard deviation and variance
• Percentile, Quantile: calculate percentiles and quantiles
• RegressionMetric: calculate one of the predefined regression metrics, such as errors of deviation from the regression line on the matrix/vector data

An example of calculating the standard deviation and percentiles for the range of bars (in points) of the current symbol and timeframe is given in the MatrixStdPercentile.mq5 file.

Log: