mcpall(ts::AbstractVector; pcut = 0.05, shuffle = 500, minlen = 3)

return vector of index of all change points in ts via recursive partitioning.
pcut is p-value threshold of existence of a change point, default is 0.05.
shuffle is number of randomizations to use,
minlen is minimum allowed length for a segment (avoid short segments).

mcplast(ts::AbstractVector; pcut=0.05, shuffle=500, minlen=3)

return index of rightmost change point in ts if any, 0 otherwise.
pcut is p-value threshold of existence of a change point, default is 0.05.
shuffle is number of randomizations to use.
minlen is minimum allowed length for a segment (avoids short segments).

mcplot(ts::AbstractVector; chgpts=Int[], palette=:seaborn_bright)

returns plot of ts with chgpts as change points (default is none),
palette is the default color scheme to use.

mcplot(ts) is a plain plot of the time series, mcplot(ts, chgpts=mcpall(ts)) is plot of ts with all change points.

mcpoint(ts::AbstractVector; pcut=0.05, shuffle=500, minlen=3)

return index of change point in mean in ts if it exists, 0 otherwise.
pcut is p-value threshold of existence of a change point, default is 0.05.
shuffle is number of randomizations to use,
minlen is minimum allowed length for a segment (prevent short segments).

mcppv(ts::AbstractVector; shuffle::Int = 500)

return p-value of a change point in mean in ts.
shuffle is number of randomization used to generate p-value, uses global random number generator.

mcptime(ts::AbstractVector; minlen::Int = 1)

return index of start of new segment via minimum ssq in ts.
minlen is minimum length of a segment you want.
0 is returned if minlen is not least twice length of ts.

ssq(v) returns sum of squares