mcovarTopN

mcovarTopN(X, Y, S, window, top, [ascending=true], [tiesMethod='oldest'])

Please see mTopN for the parameters and windowing logic.

Within a sliding window of given length (measured by the number of elements), the function stably sorts X and Y by S in the order specified by ascending, then calculates the moving covariance of the first top pairs of elements in X and Y.

The result is of type DOUBLE, with the same form as the input parameters.

Using IBM stock as an example, simulate index return rates, stock return rates, and trading volumes for six consecutive trading days:

symbol = take(`IBM, 6)
tradeDate = 2024.01.02 2024.01.03 2024.01.04 2024.01.05 2024.01.08 2024.01.09
indexRet = [0.6, 1.1, -0.2, 0.9, 1.3, 0.4]
stockRet = [0.9, 1.7, -0.1, 1.4, 1.9, 0.8]
tradeVolume = [520, 860, 610, 940, 650, 880]

stockDaily = table(symbol, tradeDate, indexRet, stockRet, tradeVolume)
stockDaily;

Output:

Over the most recent four trading days, select the top two trading days by trading volume and calculate the covariance of stockRet and indexRet:

mcovarTopN(X=indexRet, Y=stockRet, S=tradeVolume, window=4, top=2, ascending=false)
// Output: [ , 0.2, 1.17, 0.03, 0.03, 0.15]

• tradeVolume is used to select the most active trading days;
• For the data on 2024.01.09, the two trading days with the highest trading volume in the most recent four-day window correspond to the samples (0.9, 1.4) and (0.4, 0.8), whose covariance is 0.15.

Related function: mcovar