The one-number summary of a bar that the volume-weighted and money-flow
studies price on: vwap weights it by volume, moneyFlowIndex
splits it up/down, commodityChannelIndex measures its deviation and
keltner's centre line is a moving average of it. This study appends
it as a column, which is the point — those studies derive it privately,
and a caller who wants to chart it, or to run another study over it
(sma({ column: 'typicalPrice' })), needs it on the series.
Appends one column, default 'typicalPrice'. Reads high, low and
close, each named by an option defaulting to its DEFAULT_OHLCV name —
the atr shape.
Edges
No warm-up. Every bar carries its own answer, so row 0 is defined.
Length-preserving, trivially.
Linear and shift-equivariant: scaling every price by k scales the
output by k, and adding d to every price adds d to it — it is a
weighted mean of prices, so it lives in the same units. Both pinned by
property tests.
A gap in any of the three costs exactly that bar and nothing else;
there is no recursion and no window to carry it.
Nothing is clamped or ordered. Redirect high and low at columns
that cross and the mean is still their mean — a study cannot know which
of its inputs the caller meant.
Typical price —
(high + low + close) / 3. TA-Lib'sTYPPRICE, matched bar-for-bar.The one-number summary of a bar that the volume-weighted and money-flow studies price on: vwap weights it by volume, moneyFlowIndex splits it up/down, commodityChannelIndex measures its deviation and keltner's centre line is a moving average of it. This study appends it as a column, which is the point — those studies derive it privately, and a caller who wants to chart it, or to run another study over it (
sma({ column: 'typicalPrice' })), needs it on the series.Appends one column, default
'typicalPrice'. Reads high, low and close, each named by an option defaulting to itsDEFAULT_OHLCVname — the atr shape.Edges
kscales the output byk, and addingdto every price addsdto it — it is a weighted mean of prices, so it lives in the same units. Both pinned by property tests.highandlowat columns that cross and the mean is still their mean — a study cannot know which of its inputs the caller meant.