TS.FEATURES

You are here:

AXCEL.TS.FEATURES function

TS.FEATURES computes a matrix of features of the time series provided as an input.

Syntax

AXCEL.TS.FEATURES(timeseries, frequency, start, [Report])


The AXCEL.TS.FEATURES function syntax has the following arguments:

timeseries Required. The time series data in a column that you would like to decompose. Please note that your data must be a number array. It means that you should not provide a column name in your data.

frequency Required. Frequency of your data. 365 for daily, 52 for weekly, 12 for monthly, and 4 for quarterly data.

start Required. start date of your data in “yyyy-mm-dd” format.

Report Optional. By default, TS.FEATURES reports 18 features. You can define the feature names if you do not all features to be reported. For instance, you can define “Peak” to get the number of peaks, only.

when you type =AXCEL.TS.FEATURES in an Excel cell, the IntelliSense guides you through required and optional (shown in [] brackets) inputs:

In the example above, we have:

=AXCEL.TS.FEATURES(A2:A140, “2000-1-1”, 12)

This means that our data is located at cells A2 through A140 as an array (column name is not selected), it is monthly, so the. frequency is set to 12, it starts from January 1, 2000 (2000-1-1), and we would like Axcel to report all features by skipping Report for default. Here is the outcome:

As shown, Axcel reports all 18 features. Here is the list of features:

Feature Description
Trend Strength of trend
Spike measures the “spikiness” of a time series, and is computed as the variance of the leave-one-out variances of the remainder component
Linearity measures the linearity of the time series calculated based on the coefficients of orthogonal quadratic regression.
Curvature measures the curvature of a time series calculated based on the coefficients of orthogonal quadratic regression.
ACF1 the first autocorrelation coefficient of remainders ( et )
ACF10 the sum of the first ten squared autocorrelation coefficients of remainders ( et )
SeasonalImpact strength of seasonality
Peak number of peaks
Trough number of troughs
Entropy measures the “forecastability” of a time series, where low values indicate a high signal-to-noise ratio
XACF1 the autocorrelation coefficient
XACF10 the sum of the first ten squared autocorrelation coefficients
DiffACF1 the first autocorrelation coefficient of first-differenced series
DiffACF10 the sum of the first ten squared autocorrelation coefficients of first-differenced series
Diff2ACF1 the first autocorrelation coefficient of second-differenced series
Diff2ACF10 the sum of the first ten squared autocorrelation coefficients of second-differenced series
SeasonACF1 the autocorrelation coefficient at the first seasonal lag
IsUnstable the probability of being unstable based on stability and lumpiness factors