Moving averages across languages

The built-in moving averages and linear regression. Each row maps one concept from Lipi to the other languages; the word in a cell is the relation, explained in the language map overview.

Concept	Lipi	MQL4	MQL5	NinjaScript	Pine Script	thinkScript
Arnaud Legoux moving average	`ta.alma`	Absent	Absent	Absent	Caveat	Absent
Exponential moving average	`ta.ema`	Caveat	Caveat	Exact	Caveat	Caveat
Hull moving average	`ta.hma`	Absent	Absent	Differs	Differs	Differs
Wilder’s moving average	`ta.rma`	Caveat	Caveat	Absent	Caveat	Caveat
Simple moving average	`ta.sma`	Caveat	Caveat	Exact	Caveat	Caveat
Symmetrically weighted moving average	`ta.swma`	Absent	Absent	Absent	Caveat	Absent
Volume-weighted moving average	`ta.vwma`	Absent	Absent	Exact	Caveat	Absent
Weighted moving average	`ta.wma`	Caveat	Caveat	Differs	Differs	Differs
Linear regression	`ta.linreg`	Absent	Caveat	Caveat	Caveat	Caveat

Each section below gives the Lipi construct first, then the construct in each language, with a note wherever behaviour differs. The language name links to the documentation the row was checked against.

Arnaud Legoux moving average

An Arnaud Legoux moving average of source over length bars. In Lipi: ta.alma.

Language	Construct	Relation	Note
MQL4		Not in MQL4	No Arnaud Legoux average among the indicator functions; code the weights by hand.
MQL5		Not in MQL5	No ALMA among the built-in indicators; a custom indicator loaded through iCustom would have to supply it.
NinjaScript		Not in NinjaScript	No ALMA in NinjaTrader’s system indicator list; weights must be coded by hand.
Pine Script	`ta.alma`	Equivalent, with a caveat	Pine adds an optional floor flag (default false) that floors the offset calculation; Lipi’s ta.alma has no floor argument.
thinkScript		Not in thinkScript	No ALMA function or study in the functions list or the study library.

Exponential moving average

Exponential moving average of source over length bars, weighted so that recent bars count for more than older ones. In Lipi: ta.ema.

Language	Construct	Relation	Note
MQL4	`iMA with MODE_EMA`	Equivalent, with a caveat	One iMA covers four averages via ma_method (MODE_EMA = 1). ma_shift offsets the line, which Lipi lacks. Source is an applied-price constant, not any series. History indexing: shift counts bars back from the current bar; shift 0 is the still-forming bar, and any shift can be read in one call. Other timeframes: symbol and timeframe arguments read another symbol or period directly; Lipi ta calls use the chart’s own bars.
MQL5	`iMA with MODE_EMA`	Equivalent, with a caveat	Same iMA as the simple average with ma_method set to MODE_EMA. An extra ma_shift shifts the line; price is an applied-price constant. State and side effects: Returns a handle; values are read later with CopyBuffer and are not ready right after creation. History indexing: CopyBuffer start_pos 0 is the current bar and counts toward the past; copied data lands oldest-first. Other timeframes: Symbol and timeframe are explicit arguments, so another symbol or timeframe is requested directly.
NinjaScript	`EMA`	Exact
Pine Script	`ta.ema`	Equivalent, with a caveat	Pine’s reference says na source values are ignored; Lipi’s ta.ema returns na on a bar whose source is na.
thinkScript	`ExpAverage`	Equivalent, with a caveat	Same alpha, 2/(length+1), and both seed from the first value. thinkScript’s length defaults to 12, so pass it explicitly.

Hull moving average

Hull moving average of source over length bars. In Lipi: ta.hma. Built on ta.wma, so it shares ta.wma’s behaviour at this build. The half length rounds up.

Language	Construct	Relation	Note
MQL4		Not in MQL4	No Hull average among the indicator functions; build it from LWMA iMA calls.
MQL5		Not in MQL5	No Hull moving average among the built-in indicators; compose weighted averages with iMA or use iCustom.
NinjaScript	`HMA`	Looks alike, behaves differently	Lipi’s ta.hma is built on its ta.wma, which does not follow standard weights at this build, so values differ.
Pine Script	`ta.hma`	Looks alike, behaves differently	Lipi’s ta.hma uses its ta.wma, which does not follow standard weights at this build, so values differ. Its half length rounds up.
thinkScript	`HullMovingAvg`	Looks alike, behaves differently	Lipi’s ta.hma uses its ta.wma, which does not follow standard weights at this build, so values differ. The study also has a displace input.

Wilder’s moving average

Rolling moving average of source over length bars. In Lipi: ta.rma.

Language	Construct	Relation	Note
MQL4	`iMA with MODE_SMMA`	Equivalent, with a caveat	MODE_SMMA is the smoothed average; the page gives no formula. Lipi’s ta.rma uses alpha 1/length, seeded with the SMA of the first length bars. History indexing: shift counts bars back from the current bar; shift 0 is the still-forming bar, and any shift can be read in one call. Other timeframes: symbol and timeframe arguments read another symbol or period directly; Lipi ta calls use the chart’s own bars.
MQL5	`iMA with MODE_SMMA`	Equivalent, with a caveat	MODE_SMMA is the smoothed average; the page gives no formula. Lipi’s ta.rma uses alpha 1/length, seeded with the SMA of the first length bars. State and side effects: Returns a handle; values are read later with CopyBuffer and are not ready right after creation. History indexing: CopyBuffer start_pos 0 is the current bar and counts toward the past; copied data lands oldest-first. Other timeframes: Symbol and timeframe are explicit arguments, so another symbol or timeframe is requested directly.
NinjaScript		Not in NinjaScript	No RMA indicator; Wilder smoothing is used inside RSI and ATR but not exposed on its own.
Pine Script	`ta.rma`	Equivalent, with a caveat	Pine’s reference says na source values are ignored; Lipi’s ta.rma returns na on a bar whose source is na.
thinkScript	`WildersAverage`	Equivalent, with a caveat	Length defaults to 12 and seeding uses prefetched bars; Lipi needs the length and seeds with the SMA of the first length bars.

Simple moving average

Simple moving average of source over length bars: every bar in the window counts for the same. In Lipi: ta.sma.

Language	Construct	Relation	Note
MQL4	`iMA with MODE_SMA`	Equivalent, with a caveat	One iMA covers four averages via ma_method (MODE_SMA = 0). ma_shift offsets the line. Source is an applied-price constant, not any series. iMAOnArray takes arrays. History indexing: shift counts bars back from the current bar; shift 0 is the still-forming bar, and any shift can be read in one call. Other timeframes: symbol and timeframe arguments read another symbol or period directly; Lipi ta calls use the chart’s own bars.
MQL5	`iMA with MODE_SMA`	Equivalent, with a caveat	One iMA covers every average type via ma_method; MODE_SMA is simple averaging. An extra ma_shift shifts the line on the chart. State and side effects: Returns a handle; values are read later with CopyBuffer and are not ready right after creation. History indexing: CopyBuffer start_pos 0 is the current bar and counts toward the past; copied data lands oldest-first. Other timeframes: Symbol and timeframe are explicit arguments, so another symbol or timeframe is requested directly.
NinjaScript	`SMA`	Exact
Pine Script	`ta.sma`	Equivalent, with a caveat	Same average on clean data; at an na source bar Lipi returns na, TradingView returns the average of the last length non-na values.
thinkScript	`Average`	Equivalent, with a caveat	Average’s length defaults to 12, Lipi requires it, so a converter must insert 12 when omitted.

Symmetrically weighted moving average

Symmetrically weighted moving average of source over the last four bars, with the weights 1/6, 2/6, 2/6 and 1/6. In Lipi: ta.swma.

Language	Construct	Relation	Note
MQL4		Not in MQL4	No symmetric weighted average; use iMA with MODE_LWMA or weight four bars by hand.
MQL5		Not in MQL5	No symmetrically weighted average among the built-in indicators; weight four bar values yourself.
NinjaScript		Not in NinjaScript	No symmetrically weighted average in the system list; code the 1-2-2-1 weights by hand.
Pine Script	`ta.swma`	Equivalent, with a caveat	Same weights on clean data; Pine yields na for four bars after an na source value, Lipi keeps computing.
thinkScript		Not in thinkScript	No symmetrically weighted average; WMA weights linearly over a length you choose.

Volume-weighted moving average

Volume-weighted moving average of source over length bars: bars that traded more count for more. In Lipi: ta.vwma.

Language	Construct	Relation	Note
MQL4		Not in MQL4	No volume-weighted moving average among the indicator functions.
MQL5		Not in MQL5	No volume-weighted moving average among the built-in indicators; iMA has no volume weighting.
NinjaScript	`VWMA`	Exact
Pine Script	`ta.vwma`	Equivalent, with a caveat	Pine’s reference says na source values are ignored; Lipi’s ta.vwma returns na on a bar whose source is na.
thinkScript		Not in thinkScript	No volume-weighted moving average; build it as Sum(price*volume, n) divided by Sum(volume, n).

Weighted moving average

Weighted moving average of source over length bars. In Lipi: ta.wma. At this build the result does not follow standard WMA weights: a step from 0 to 1 settles at 1/length of the step.

Language	Construct	Relation	Note
MQL4	`iMA with MODE_LWMA`	Equivalent, with a caveat	One iMA covers four averages via ma_method (MODE_LWMA = 3). ma_shift offsets the line. Source is an applied-price constant, not any series. History indexing: shift counts bars back from the current bar; shift 0 is the still-forming bar, and any shift can be read in one call. Other timeframes: symbol and timeframe arguments read another symbol or period directly; Lipi ta calls use the chart’s own bars.
MQL5	`iMA with MODE_LWMA`	Equivalent, with a caveat	MODE_LWMA is linear-weighted averaging, set through ma_method on iMA. An extra ma_shift shifts the line on the chart. State and side effects: Returns a handle; values are read later with CopyBuffer and are not ready right after creation. History indexing: CopyBuffer start_pos 0 is the current bar and counts toward the past; copied data lands oldest-first. Other timeframes: Symbol and timeframe are explicit arguments, so another symbol or timeframe is requested directly.
NinjaScript	`WMA`	Looks alike, behaves differently	At this build Lipi’s ta.wma does not follow standard weights: a step from 0 to 1 settles at 1/length. Compare values before porting.
Pine Script	`ta.wma`	Looks alike, behaves differently	At this build Lipi’s ta.wma does not follow standard weights: a step from 0 to 1 settles at 1/length. Compare values before porting.
thinkScript	`WMA`	Looks alike, behaves differently	WMA’s length defaults to 9. At this build Lipi’s ta.wma does not follow standard weights: a step from 0 to 1 settles at 1/length.

Linear regression

Fits a linear regression to source across length bars and returns a value from that fit. In Lipi: ta.linreg. At this build a non-zero offset gives values that do not follow the fitted line.

Language	Construct	Relation	Note
MQL4		Not in MQL4	No linear-regression function among the indicator functions.
MQL5	`vector::LinearRegression`	Equivalent, with a caveat	Returns the whole fitted line for a vector; pick the element you need. No offset argument; fill the vector with the window first.
NinjaScript	`LinReg`	Equivalent, with a caveat	No offset argument; the call gives the least-squares line value. Page states no formula or which point of the fit is returned.
Pine Script	`ta.linreg`	Equivalent, with a caveat	Matches the page’s fit at offset 0; at other offsets Lipi’s values do not follow the fit at this build, and Lipi resumes after na.
thinkScript	`Inertia`	Equivalent, with a caveat	Least-squares regression value at the current bar; no offset argument and no default length, unlike Lipi ta.linreg’s offset.

Manual

Use Lipi with AI

Reference

Moving averages across languages

Arnaud Legoux moving average

Exponential moving average

Hull moving average

Wilder’s moving average

Simple moving average

Symmetrically weighted moving average

Volume-weighted moving average

Weighted moving average

Linear regression

​Arnaud Legoux moving average

​Exponential moving average

​Hull moving average

​Wilder’s moving average

​Simple moving average

​Symmetrically weighted moving average

​Volume-weighted moving average

​Weighted moving average

​Linear regression

Arnaud Legoux moving average

Exponential moving average

Hull moving average

Wilder’s moving average

Simple moving average

Symmetrically weighted moving average

Volume-weighted moving average

Weighted moving average

Linear regression