# Fitting

Summary: this tutorial will walk through several ways of fitting data with FSharp.Stats.

## Linear Regression

In Linear Regression a linear system of equations is generated. The coefficients obtained by the solution to this equation system minimize the squared distances from the regression curve to the data points. These distances are also known as residuals (or least squares).

### Simple Linear Regression

Simple linear regression aims to fit a straight regression line to the data. While the least squares approach efficiently minimizes the sum of squared residuals it is prone to outliers. An alternative is a robust simple linear regression like Theil's incomplete method or the Theil-Sen estimator, that are outlier resistant.

#### Univariable

open FSharp.Stats
open FSharp.Stats.Fitting.LinearRegression

let xData = vector [|1. .. 10.|]
let yData = vector [|4.;7.;9.;12.;15.;17.;16.;23.;5.;30.|]

//Least squares simple linear regression
let coefficientsLinearLS =
OrdinaryLeastSquares.Linear.Univariable.coefficient xData yData
let fittingFunctionLinearLS x =
OrdinaryLeastSquares.Linear.Univariable.fit coefficientsLinearLS x

//Robust simple linear regression
let coefficientsLinearRobust =
RobustRegression.Linear.theilSenEstimator xData yData
let fittingFunctionLinearRobust x =
RobustRegression.Linear.fit coefficientsLinearRobust x

//least squares simple linear regression through the origin
let coefficientsLinearRTO =
OrdinaryLeastSquares.Linear.RTO.coefficientOfVector xData yData
let fittingFunctionLinearRTO x =
OrdinaryLeastSquares.Linear.RTO.fit coefficientsLinearRTO x

open Plotly.NET

//some axis styling
let myAxis title = Axis.LinearAxis.init(Title=title,Mirror=StyleParam.Mirror.All,Ticks=StyleParam.TickOptions.Inside,Showgrid=false,Showline=true,Zeroline=false)
let styleChart xt yt c = c |> Chart.withX_Axis (myAxis xt) |> Chart.withY_Axis (myAxis yt)

let rawChart =
Chart.Point(xData,yData)
|> Chart.withTraceName "raw data"

let fittingLS =
let fit =
[|0. .. 11.|]
|> Array.map (fun x -> x,fittingFunctionLinearLS x)
Chart.Line(fit)
|> Chart.withTraceName "least squares (LS)"

let fittingRobust =
let fit =
[|0. .. 11.|]
|> Array.map (fun x -> x,fittingFunctionLinearRobust x)
Chart.Line(fit)
|> Chart.withTraceName "TheilSen estimator"

let fittingRTO =
let fit =
[|0. .. 11.|]
|> Array.map (fun x -> x,fittingFunctionLinearRTO x)
Chart.Line(fit)
|> Chart.withTraceName "LS through origin"

let simpleLinearChart =
[rawChart;fittingLS;fittingRTO;fittingRobust;]
|> Chart.Combine
|> styleChart "" ""


#### Multivariable

//Multivariate simple linear regression
let xVectorMulti =
[
[1.; 1. ;2.  ]
[2.; 0.5;6.  ]
[3.; 0.8;10. ]
[4.; 2. ;14. ]
[5.; 4. ;18. ]
[6.; 3. ;22. ]
]
|> Matrix.ofJaggedSeq

let yVectorMulti =
let transformX (x:Matrix<float>) =
x
|> Matrix.mapiRows (fun _ v -> 100. + (v.[0] * 2.5) + (v.[1] * 4.) + (v.[2] * 0.5))
xVectorMulti
|> transformX
|> vector

let coefficientsMV =
OrdinaryLeastSquares.Linear.Multivariable.coefficients xVectorMulti yVectorMulti
let fittingFunctionMV x =
OrdinaryLeastSquares.Linear.Multivariable.fit coefficientsMV x


### Polynomial Regression

In polynomial regression a higher degree (d > 1) polynomial is fitted to the data. The coefficients are chosen that the sum of squared residuals is minimized.

open FSharp.Stats
open FSharp.Stats.Fitting.LinearRegression

let xDataP = vector [|1. .. 10.|]
let yDataP = vector [|4.;7.;9.;8.;6.;3.;2.;5.;6.;8.;|]

//Least squares polynomial regression

//define the order the polynomial should have (order 3: f(x) = ax^3 + bx^2 + cx + d)
let order = 3
let coefficientsPol =
OrdinaryLeastSquares.Polynomial.coefficient order xDataP yDataP
let fittingFunctionPol x =
OrdinaryLeastSquares.Polynomial.fit order coefficientsPol x

//weighted least squares polynomial regression
//If heteroscedasticity is assumed or the impact of single datapoints should be
//increased/decreased you can use a weighted version of the polynomial regression.

//define the order the polynomial should have (order 3: f(x) = ax^3 + bx^2 + cx + d)
let orderP = 3

//define the weighting vector
let weights = yDataP |> Vector.map (fun y -> 1. / y)
let coefficientsPolW =
OrdinaryLeastSquares.Polynomial.coefficientsWithWeighting orderP weights xDataP yDataP
let fittingFunctionPolW x =
OrdinaryLeastSquares.Polynomial.fit orderP coefficientsPolW x

let rawChartP =
Chart.Point(xDataP,yDataP)
|> Chart.withTraceName "raw data"

let fittingPol =
let fit =
[|1. .. 0.1 .. 10.|]
|> Array.map (fun x -> x,fittingFunctionPol x)
Chart.Line(fit)
|> Chart.withTraceName "order = 3"

let fittingPolW =
let fit =
[|1. .. 0.1 .. 10.|]
|> Array.map (fun x -> x,fittingFunctionPolW x)
Chart.Line(fit)
|> Chart.withTraceName "order = 3 weigthed"

let polRegressionChart =
[rawChartP;fittingPol;fittingPolW]
|> Chart.Combine
|> styleChart "" ""


## Nonlinear Regression

Nonlinear Regression is used if a known model should be fitted to the data that cannot be represented in a linear system of equations. Common examples are:

• gaussian functions
• log functions
• exponential functions

To fit such models to your data the NonLinearRegression module can be used. Three solver-methods are available to iteratively converge to a minimal least squares value.

• GaussNewton
• LevenbergMarquardt
• LevenbergMarquardtConstrained

For solving a nonlinear problem the model function has to be converted to a NonLinearRegression.Model type consisting of

• parameter names,
• the function itself, and
• partial derivatives of all unknown parameters.

For clarification a exponential relationship in the form of y = a * exp(b * x) should be solved:

open System
open FSharp.Stats.Fitting
open FSharp.Stats.Fitting.LinearRegression
open FSharp.Stats.Fitting.NonLinearRegression

let xDataN = [|1.;2.; 3.; 4.|]
let yDataN = [|5.;14.;65.;100.|]

//search for:  y = a * exp(b * x)

// 1. create the model
// 1.1 define parameter names
let parameterNames = [|"a";"b"|]

// 1.2 define the exponential function that gets a parameter vector containing the
//searched parameters and the x value and gives the corresponding y value
let getFunctionValue =
fun (parameterVector: Vector<float>) x ->
parameterVector.[0] * Math.Exp(parameterVector.[1] * x)
//a                   *      exp(b                   * x)

// 1.3 Define partial derivatives of the exponential function.
//     Take partial derivatives for every unknown parameter and
//     insert it into the gradient vector sorted by parameterNames.
fun (parameterVector:Vector<float>) (gradientVector: Vector<float>) xValueN ->
// partial derivative of y=a*exp(b*x) in respect to the first parameter (a)   --> exp(b*x)
// partial derivative of y=a*exp(b*x) in respect to the second parameter (b)  --> a*x*exp(b*x)
gradientVector.[1] <- parameterVector.[0] * xValueN * Math.Exp(parameterVector.[1] * xValueN)

// 1.4 create the model
let model = createModel parameterNames getFunctionValue getGradientValues

// 2. define the solver options
// 2.1 define the stepwidth of the x value change
let deltaX = 0.0001

// 2.2 define the stepwidth of the parameter change
let deltaP = 0.0001

// 2.3 define the number of iterations
let k = 1000

// 2.4 define an initial guess
//     For many problems you can set a default value or let the user decide to choose an
//     appropriate guess. In the case of an exponential or log model you can use the
//     solution of the linearized problem as a first guess.
let initialParamGuess (xData:float []) (yData:float [])=
//gets the linear representation of the problem and solves it by simple linear regression
//(prone to least-squares-deviations at high y_Values)
let yLn = yData |> Array.map (fun x -> Math.Log(x)) |> vector
let linearReg =
LinearRegression.OrdinaryLeastSquares.Linear.Univariable.coefficient (vector xData) yLn
//calculates the parameters back into the exponential representation
let a = exp linearReg.[0]
let b = linearReg.[1]
[|a;b|]

// 2.5 create the solverOptions
let solverOptions =
let guess = initialParamGuess xDataN yDataN
NonLinearRegression.createSolverOption 0.0001 0.0001 1000 guess

// 3. get coefficients
let coefficientsExp = GaussNewton.estimatedParams model solverOptions xDataN yDataN
//val coefficients = vector [|5.68867298; 0.7263428835|]

// 4. create fitting function
let fittingFunction x = coefficientsExp.[0] * Math.Exp(coefficientsExp.[1] * x)

let rawChartNLR =
Chart.Point(xDataN,yDataN)
|> Chart.withTraceName "raw data"

let fittingNLR =
let fit =
[|1. .. 0.1 .. 10.|]
|> Array.map (fun x -> x,fittingFunction x)
Chart.Line(fit)
|> Chart.withTraceName "NLR"

let NLRChart =
[rawChartNLR;fittingNLR]
|> Chart.Combine
|> styleChart "" ""


### LevenbergMarquardtConstrained

For nonlinear regression using the LevenbergMarquardtConstrained module, you have to follow similar steps as in the example shown above. In this example, a logistic function of the form y = L/(1+e^(-k(t-x))) should be fitted to count data:

open FSharp.Stats.Fitting.NonLinearRegression

let xHours = [|0.; 19.5; 25.5; 30.; 43.; 48.5; 67.75|]

let yCount = [|0.0; 2510000.0; 4926400.0; 9802600.0; 14949400.0; 15598800.0; 16382000.0|]

// 1. Choose a model
// The model we need already exists in FSharp.Stats and can be taken from the "Table" module.
let model' = Table.LogisticFunctionAscending

// 2. Define the solver options
// 2.1 Initial parameter guess
// The solver needs an initial parameter guess. This can be done by the user or with an estimator function.
// The cutoffPercentage says at which percentage of the y-Range the lower part of the slope is.
// Manual curation of parameter guesses can be performed in this step by editing the param array.
let initialParamGuess' = LevenbergMarquardtConstrained.initialParam xHours yCount 0.1

// 2.2 Create the solver options
let solverOptions' = Table.lineSolverOptions initialParamGuess'

// 3. Estimate parameters for a possible solution based on residual sum of squares
// Besides the solverOptions, an upper and lower bound for the parameters are required.
// It is recommended to define them depending on the initial param guess
let lowerBound =
initialParamGuess'
|> Array.map (fun param ->
// Initial paramters -20%
param - (abs param) * 0.2
)
|> vector

let upperBound =
initialParamGuess'
|> Array.map (fun param ->
param + (abs param) * 0.2
)
|> vector

let estParams =
LevenbergMarquardtConstrained.estimatedParams model' solverOptions' 0.001 10. lowerBound upperBound xHours yCount

// 3.1 Estimate multiple parameters and pick the one with the least residual sum of squares (RSS)
// For a more "global" minimum. it is possible to estimate multiple possible parameters for different initial guesses.

//3.1.1 Generation of parameters with varying steepnesses
let multipleSolverOptions =
LevenbergMarquardtConstrained.initialParamsOverRange xHours yCount [|0. .. 0.1 .. 2.|]
|> Array.map Table.lineSolverOptions

multipleSolverOptions
|> Array.map (fun solvO ->
let lowerBound =
solvO.InitialParamGuess
|> Array.map (fun param -> param - (abs param) * 0.2)
|> vector
let upperBound =
solvO.InitialParamGuess
|> Array.map (fun param -> param + (abs param) * 0.2)
|> vector
model' solvO 0.001 10. lowerBound upperBound xHours yCount
)
|> Array.minBy snd
|> fst

// The result is the same as for 'estParams', but tupled with the corresponding RSS, which can be taken
// as a measure of quality for the estimate.

// 4. Create fitting function

let fittedY = Array.zip [|1. .. 68.|] ([|1. .. 68.|] |> Array.map fittingFunction')

let fittedLogisticFunc =
[
Chart.Point (Array.zip xHours yCount)
|> Chart.withTraceName"Data Points"
Chart.Line fittedY
|> Chart.withTraceName "Fit"
]
|> Chart.Combine
|> styleChart "Time" "Count"


## Smoothing spline

A smoothing spline aims to minimize a function consisting of two error terms:

• error1: sum of squared residuals

• Similar to the OrdinaryLeastSquares regression this error term ensures the fidelity to the data.
• error2: integral of the second derivative of the fitting function

• This error term ensures the smoothness of the resulting curve.

A smoothing parameter (lambda) mediates between the two error terms.

• E = error1 + (lambda * error2)

• If lambda = 0, the resulting curve minimizes the sum of squared residuals and results in an interpolating curve.
• If lambda = infinity, the resulting curve is punished by the smoothness measurement and results in a straight regression line.

The spline is constructed out of piecewise cubic polynomials that meet at knots. In the defined knots the function is continuous. Depending on the used smoothing factor and the defined knots the smoothing spline has a unique solution. The resulting curve is just defined within the interval defined in the x values of the data.

The right amount of smoothing can be determined by cross validation or generalized cross validation.

open FSharp.Stats.Fitting

let xDataS = [|1.;2.; 3.; 4.|]
let yDataS = [|5.;14.;65.;75.|]

let data = Array.zip xDataS yDataS

//in every x position a knot should be located
let knots = xDataS

let spline lambda x = (Spline.smoothingSpline data knots) lambda x

let fit lambda =
[|1. .. 0.1 .. 4.|]
|> Array.map (fun x -> x,spline lambda x)
|> Chart.Line
|> Chart.withTraceName (sprintf "lambda: %.3f" lambda)

let rawChartS = Chart.Point(data)

let smoothingSplines =
[
rawChartS
fit 0.001
fit 0.02
fit 1.
]
|> Chart.Combine
|> styleChart "" ""

Multiple items
namespace FSharp

--------------------
namespace Microsoft.FSharp
namespace FSharp.Stats
namespace FSharp.Stats.Fitting
module LinearRegression

from FSharp.Stats.Fitting
val xData : Vector<float>
Multiple items
val vector : l:seq<float> -> Vector<float>

--------------------
type vector = Vector<float>
val yData : Vector<float>
val coefficientsLinearLS : Vector<float>
module OrdinaryLeastSquares

from FSharp.Stats.Fitting.LinearRegression
module Linear

from FSharp.Stats.Fitting.LinearRegression.OrdinaryLeastSquares
module Univariable

from FSharp.Stats.Fitting.LinearRegression.OrdinaryLeastSquares.Linear
val coefficient : xData:Vector<float> -> yData:Vector<float> -> Vector<float>
val fittingFunctionLinearLS : x:float -> float
val x : float
val fit : coef:Vector<float> -> x:float -> float
val coefficientsLinearRobust : Vector<float>
module RobustRegression

from FSharp.Stats.Fitting.LinearRegression
module Linear

from FSharp.Stats.Fitting.LinearRegression.RobustRegression
val theilSenEstimator : xData:Vector<float> -> yData:Vector<float> -> Vector<float>
val fittingFunctionLinearRobust : x:float -> float
val fit : (Vector<float> -> float -> float)
val coefficientsLinearRTO : float
module RTO

from FSharp.Stats.Fitting.LinearRegression.OrdinaryLeastSquares.Linear
val coefficientOfVector : x:Vector<float> -> y:Vector<float> -> float
val fittingFunctionLinearRTO : x:float -> float
val fit : coef:float -> x:float -> float
namespace Plotly
namespace Plotly.NET
val myAxis : title:string -> Axis.LinearAxis
val title : string
module Axis

from Plotly.NET
Multiple items
type LinearAxis =
inherit DynamicObj
new : unit -> LinearAxis
static member init : ?AxisType:AxisType * ?Title:string * ?Titlefont:Font * ?Autorange:AutoRange * ?Rangemode:RangeMode * ?Range:Range * ?RangeSlider:RangeSlider * ?Fixedrange:'a * ?Tickmode:TickMode * ?nTicks:'b * ?Tick0:'c * ?dTick:'d * ?Tickvals:'e * ?Ticktext:'f * ?Ticks:TickOptions * ?Mirror:Mirror * ?Ticklen:'g * ?Tickwidth:'h * ?Tickcolor:'i * ?Showticklabels:'j * ?Tickfont:Font * ?Tickangle:'k * ?Tickprefix:'l * ?Showtickprefix:ShowTickOption * ?Ticksuffix:'m * ?Showticksuffix:ShowTickOption * ?Showexponent:ShowExponent * ?Exponentformat:ExponentFormat * ?Tickformat:'n * ?Hoverformat:'o * ?Showline:bool * ?Linecolor:'p * ?Linewidth:'q * ?Showgrid:bool * ?Gridcolor:'r * ?Gridwidth:'s * ?Zeroline:bool * ?Zerolinecolor:'t * ?Zerolinewidth:'a1 * ?Anchor:AxisAnchorId * ?Side:Side * ?Overlaying:AxisAnchorId * ?Domain:Range * ?Position:float * ?IsSubplotObj:'a2 * ?Tickvalssrc:'a3 * ?Ticktextsrc:'a4 * ?Showspikes:'a5 * ?Spikesides:'a6 * ?Spikethickness:'a7 * ?Spikecolor:'a8 * ?Showbackground:'a9 * ?Backgroundcolor:'a10 * ?Showaxeslabels:'a11 -> LinearAxis
static member style : ?AxisType:AxisType * ?Title:string * ?Titlefont:Font * ?Autorange:AutoRange * ?Rangemode:RangeMode * ?Range:Range * ?RangeSlider:RangeSlider * ?Fixedrange:'a * ?Tickmode:TickMode * ?nTicks:'b * ?Tick0:'c * ?dTick:'d * ?Tickvals:'e * ?Ticktext:'f * ?Ticks:TickOptions * ?Mirror:Mirror * ?Ticklen:'g * ?Tickwidth:'h * ?Tickcolor:'i * ?Showticklabels:'j * ?Tickfont:Font * ?Tickangle:'k * ?Tickprefix:'l * ?Showtickprefix:ShowTickOption * ?Ticksuffix:'m * ?Showticksuffix:ShowTickOption * ?Showexponent:ShowExponent * ?Exponentformat:ExponentFormat * ?Tickformat:'n * ?Hoverformat:'o * ?Showline:bool * ?Linecolor:'p * ?Linewidth:'q * ?Showgrid:bool * ?Gridcolor:'r * ?Gridwidth:'s * ?Zeroline:bool * ?Zerolinecolor:'t * ?Zerolinewidth:'a1 * ?Anchor:AxisAnchorId * ?Side:Side * ?Overlaying:AxisAnchorId * ?Domain:Range * ?Position:float * ?IsSubplotObj:'a2 * ?Tickvalssrc:'a3 * ?Ticktextsrc:'a4 * ?Showspikes:'a5 * ?Spikesides:'a6 * ?Spikethickness:'a7 * ?Spikecolor:'a8 * ?Showbackground:'a9 * ?Backgroundcolor:'a10 * ?Showaxeslabels:'a11 -> (LinearAxis -> LinearAxis)

--------------------
new : unit -> Axis.LinearAxis
static member Axis.LinearAxis.init : ?AxisType:StyleParam.AxisType * ?Title:string * ?Titlefont:Font * ?Autorange:StyleParam.AutoRange * ?Rangemode:StyleParam.RangeMode * ?Range:StyleParam.Range * ?RangeSlider:RangeSlider * ?Fixedrange:'a * ?Tickmode:StyleParam.TickMode * ?nTicks:'b * ?Tick0:'c * ?dTick:'d * ?Tickvals:'e * ?Ticktext:'f * ?Ticks:StyleParam.TickOptions * ?Mirror:StyleParam.Mirror * ?Ticklen:'g * ?Tickwidth:'h * ?Tickcolor:'i * ?Showticklabels:'j * ?Tickfont:Font * ?Tickangle:'k * ?Tickprefix:'l * ?Showtickprefix:StyleParam.ShowTickOption * ?Ticksuffix:'m * ?Showticksuffix:StyleParam.ShowTickOption * ?Showexponent:StyleParam.ShowExponent * ?Exponentformat:StyleParam.ExponentFormat * ?Tickformat:'n * ?Hoverformat:'o * ?Showline:bool * ?Linecolor:'p * ?Linewidth:'q * ?Showgrid:bool * ?Gridcolor:'r * ?Gridwidth:'s * ?Zeroline:bool * ?Zerolinecolor:'t * ?Zerolinewidth:'a1 * ?Anchor:StyleParam.AxisAnchorId * ?Side:StyleParam.Side * ?Overlaying:StyleParam.AxisAnchorId * ?Domain:StyleParam.Range * ?Position:float * ?IsSubplotObj:'a2 * ?Tickvalssrc:'a3 * ?Ticktextsrc:'a4 * ?Showspikes:'a5 * ?Spikesides:'a6 * ?Spikethickness:'a7 * ?Spikecolor:'a8 * ?Showbackground:'a9 * ?Backgroundcolor:'a10 * ?Showaxeslabels:'a11 -> Axis.LinearAxis
module StyleParam

from Plotly.NET
type Mirror =
| True
| Ticks
| False
| All
| AllTicks
static member convert : (Mirror -> obj)
static member toString : (Mirror -> string)
union case StyleParam.Mirror.All: StyleParam.Mirror
type TickOptions =
| Outside
| Inside
| Empty
static member convert : (TickOptions -> obj)
static member toString : (TickOptions -> string)
union case StyleParam.TickOptions.Inside: StyleParam.TickOptions
val styleChart : xt:string -> yt:string -> c:GenericChart.GenericChart -> GenericChart.GenericChart
val xt : string
val yt : string
val c : GenericChart.GenericChart
type Chart =
static member Area : xy:seq<#IConvertible * #IConvertible> * ?Name:string * ?ShowMarkers:bool * ?Showlegend:bool * ?MarkerSymbol:Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:TextPosition * ?TextFont:Font * ?Dash:DrawingStyle * ?Width:'a2 -> GenericChart
static member Area : x:seq<#IConvertible> * y:seq<#IConvertible> * ?Name:string * ?ShowMarkers:bool * ?Showlegend:bool * ?MarkerSymbol:Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:TextPosition * ?TextFont:Font * ?Dash:DrawingStyle * ?Width:'a2 -> GenericChart
static member Bar : keysvalues:seq<#IConvertible * #IConvertible> * ?Name:string * ?Showlegend:bool * ?Color:'a2 * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:TextPosition * ?TextFont:Font * ?Marker:Marker -> GenericChart
static member Bar : keys:seq<#IConvertible> * values:seq<#IConvertible> * ?Name:string * ?Showlegend:bool * ?Color:'a2 * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:TextPosition * ?TextFont:Font * ?Marker:Marker -> GenericChart
static member BoxPlot : xy:seq<'a0 * 'a1> * ?Name:string * ?Showlegend:bool * ?Color:string * ?Fillcolor:'a2 * ?Opacity:float * ?Whiskerwidth:'a3 * ?Boxpoints:Boxpoints * ?Boxmean:BoxMean * ?Jitter:'a4 * ?Pointpos:'a5 * ?Orientation:Orientation * ?Marker:Marker * ?Line:Line * ?Alignmentgroup:'a6 * ?Offsetgroup:'a7 * ?Notched:bool * ?NotchWidth:float * ?QuartileMethod:QuartileMethod -> GenericChart
static member BoxPlot : ?x:'a0 * ?y:'a1 * ?Name:string * ?Showlegend:bool * ?Color:string * ?Fillcolor:'a2 * ?Opacity:float * ?Whiskerwidth:'a3 * ?Boxpoints:Boxpoints * ?Boxmean:BoxMean * ?Jitter:'a4 * ?Pointpos:'a5 * ?Orientation:Orientation * ?Marker:Marker * ?Line:Line * ?Alignmentgroup:'a6 * ?Offsetgroup:'a7 * ?Notched:bool * ?NotchWidth:float * ?QuartileMethod:QuartileMethod -> GenericChart
static member Bubble : xysizes:seq<#IConvertible * #IConvertible * #IConvertible> * ?Name:string * ?Showlegend:bool * ?MarkerSymbol:Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:TextPosition * ?TextFont:Font * ?StackGroup:string * ?Orientation:Orientation * ?GroupNorm:GroupNorm * ?UseWebGL:bool -> GenericChart
static member Bubble : x:seq<#IConvertible> * y:seq<#IConvertible> * sizes:seq<#IConvertible> * ?Name:string * ?Showlegend:bool * ?MarkerSymbol:Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:TextPosition * ?TextFont:Font * ?StackGroup:string * ?Orientation:Orientation * ?GroupNorm:GroupNorm * ?UseWebGL:bool -> GenericChart
static member Candlestick : stockTimeSeries:seq<DateTime * StockData> * ?Increasing:Line * ?Decreasing:Line * ?WhiskerWidth:float * ?Line:Line * ?XCalendar:Calendar -> GenericChart
static member Candlestick : open:seq<#IConvertible> * high:seq<#IConvertible> * low:seq<#IConvertible> * close:seq<#IConvertible> * x:seq<#IConvertible> * ?Increasing:Line * ?Decreasing:Line * ?WhiskerWidth:float * ?Line:Line * ?XCalendar:Calendar -> GenericChart
...
static member Chart.withX_Axis : xAxis:Axis.LinearAxis * ?Id:int -> (GenericChart.GenericChart -> GenericChart.GenericChart)
static member Chart.withY_Axis : yAxis:Axis.LinearAxis * ?Id:int -> (GenericChart.GenericChart -> GenericChart.GenericChart)
val rawChart : GenericChart.GenericChart
static member Chart.Point : xy:seq<#System.IConvertible * #System.IConvertible> * ?Name:string * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
static member Chart.Point : x:seq<#System.IConvertible> * y:seq<#System.IConvertible> * ?Name:string * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
static member Chart.withTraceName : ?Name:string * ?Showlegend:bool * ?Legendgroup:string * ?Visible:StyleParam.Visible -> (GenericChart.GenericChart -> GenericChart.GenericChart)
val fittingLS : GenericChart.GenericChart
val fit : (float * float) []
Multiple items
module Array

from FSharp.Stats

--------------------
module Array

from Microsoft.FSharp.Collections
val map : mapping:('T -> 'U) -> array:'T [] -> 'U []
static member Chart.Line : xy:seq<#System.IConvertible * #System.IConvertible> * ?Name:string * ?ShowMarkers:bool * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?Dash:'c * ?Width:'d * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
static member Chart.Line : x:seq<#System.IConvertible> * y:seq<#System.IConvertible> * ?Name:string * ?ShowMarkers:bool * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?Dash:'a2 * ?Width:'a3 * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
val fittingRobust : GenericChart.GenericChart
val fittingRTO : GenericChart.GenericChart
val simpleLinearChart : GenericChart.GenericChart
static member Chart.Combine : gCharts:seq<GenericChart.GenericChart> -> GenericChart.GenericChart
module GenericChart

from Plotly.NET
val toChartHTML : gChart:GenericChart.GenericChart -> string
val xVectorMulti : Matrix<float>
Multiple items
module Matrix

from FSharp.Stats

--------------------
type Matrix<'T> =
| DenseRepr of DenseMatrix<'T>
| SparseRepr of SparseMatrix<'T>
interface IEnumerable
interface IEnumerable<'T>
interface IStructuralEquatable
interface IStructuralComparable
interface IComparable
override Equals : yobj:obj -> bool
override GetHashCode : unit -> int
member GetSlice : start1:int option * finish1:int option * start2:int option * finish2:int option -> Matrix<'T>
member PermuteColumns : p:permutation -> Matrix<'T>
member PermuteRows : p:permutation -> Matrix<'T>
...
val ofJaggedSeq : xss:seq<#seq<float>> -> Matrix<float>
val yVectorMulti : Vector<float>
val transformX : (Matrix<float> -> seq<float>)
val x : Matrix<float>
Multiple items
val float : value:'T -> float (requires member op_Explicit)

--------------------
type float = System.Double

--------------------
type float<'Measure> = float
val mapiRows : f:(int -> rowvec -> 'b) -> m:matrix -> seq<'b>
val v : rowvec
val coefficientsMV : Vector<float>
module Multivariable

from FSharp.Stats.Fitting.LinearRegression.OrdinaryLeastSquares.Linear
val coefficients : xData:Matrix<float> -> yData:Vector<float> -> Vector<float>
val fittingFunctionMV : x:Vector<float> -> float
val x : Vector<float>
val fit : coef:Vector<float> -> x:Vector<float> -> float
val xDataP : Vector<float>
val yDataP : Vector<float>
val order : int
val coefficientsPol : Vector<float>
module Polynomial

from FSharp.Stats.Fitting.LinearRegression.OrdinaryLeastSquares
val coefficient : order:int -> xData:Vector<float> -> yData:Vector<float> -> Vector<float>
val fittingFunctionPol : x:float -> float
val fit : order:int -> coef:Vector<float> -> x:float -> float
val orderP : int
val weights : Vector<float>
Multiple items
module Vector

from FSharp.Stats

--------------------
type Vector<'T> =
interface IEnumerable
interface IEnumerable<'T>
interface IStructuralEquatable
interface IStructuralComparable
interface IComparable
new : opsV:INumeric<'T> option * arrV:'T array -> Vector<'T>
override Equals : yobj:obj -> bool
override GetHashCode : unit -> int
member GetSlice : start:int option * finish:int option -> Vector<'T>
member Permute : p:permutation -> Vector<'T>
...

--------------------
new : opsV:INumeric<'T> option * arrV:'T array -> Vector<'T>
val map : mapping:(float -> float) -> vector:vector -> Vector<float>
val y : float
val coefficientsPolW : Vector<float>
val coefficientsWithWeighting : order:int -> weighting:Vector<float> -> xData:Vector<float> -> yData:Vector<float> -> Vector<float>
val fittingFunctionPolW : x:float -> float
val rawChartP : GenericChart.GenericChart
val fittingPol : GenericChart.GenericChart
val fittingPolW : GenericChart.GenericChart
val polRegressionChart : GenericChart.GenericChart
namespace System
module NonLinearRegression

from FSharp.Stats.Fitting
val xDataN : float []
val yDataN : float []
val parameterNames : string []
val getFunctionValue : parameterVector:Vector<float> -> x:float -> float
val parameterVector : Vector<float>
Multiple items
val float : value:'T -> float (requires member op_Explicit)

--------------------
type float = Double

--------------------
type float<'Measure> = float
type Math =
static val E : float
static val PI : float
static val Tau : float
static member Abs : value:decimal -> decimal + 6 overloads
static member Acos : d:float -> float
static member Acosh : d:float -> float
static member Asin : d:float -> float
static member Asinh : d:float -> float
static member Atan : d:float -> float
static member Atan2 : y:float * x:float -> float
...
Math.Exp(d: float) : float
val xValueN : float
val model : Model
val createModel : parameterNames:string [] -> getFunctionValue:(Vector<float> -> float -> float) -> getGradientValue:(Vector<float> -> Vector<float> -> float -> Vector<float>) -> Model
val deltaX : float
val deltaP : float
val k : int
val initialParamGuess : xData:float [] -> yData:float [] -> float []
val xData : float []
val yData : float []
val yLn : Vector<float>
type Array =
member Clone : unit -> obj
member CopyTo : array:Array * index:int -> unit + 1 overload
member GetEnumerator : unit -> IEnumerator
member GetLength : dimension:int -> int
member GetLongLength : dimension:int -> int64
member GetLowerBound : dimension:int -> int
member GetUpperBound : dimension:int -> int
member GetValue : index:int -> obj + 7 overloads
member Initialize : unit -> unit
member IsFixedSize : bool
...
Math.Log(d: float) : float
Math.Log(a: float, newBase: float) : float
val linearReg : Vector<float>
val a : float
val exp : value:'T -> 'T (requires member Exp)
val b : float
val solverOptions : SolverOptions
val guess : float []
val createSolverOption : minimumDeltaValue:float -> minimumDeltaParameters:float -> maximumIterations:int -> initialParamGuess:float [] -> SolverOptions
val coefficientsExp : vector
module GaussNewton

from FSharp.Stats.Fitting.NonLinearRegression
val estimatedParams : model:Model -> solverOptions:SolverOptions -> xData:float [] -> yData:float [] -> vector
val fittingFunction : x:float -> float
val rawChartNLR : GenericChart.GenericChart
static member Chart.Point : xy:seq<#IConvertible * #IConvertible> * ?Name:string * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
static member Chart.Point : x:seq<#IConvertible> * y:seq<#IConvertible> * ?Name:string * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
val fittingNLR : GenericChart.GenericChart
static member Chart.Line : xy:seq<#IConvertible * #IConvertible> * ?Name:string * ?ShowMarkers:bool * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?Dash:'c * ?Width:'d * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
static member Chart.Line : x:seq<#IConvertible> * y:seq<#IConvertible> * ?Name:string * ?ShowMarkers:bool * ?Showlegend:bool * ?MarkerSymbol:StyleParam.Symbol * ?Color:string * ?Opacity:float * ?Labels:seq<string> * ?TextPosition:StyleParam.TextPosition * ?TextFont:Font * ?Dash:'a2 * ?Width:'a3 * ?StackGroup:string * ?Orientation:StyleParam.Orientation * ?GroupNorm:StyleParam.GroupNorm * ?UseWebGL:bool -> GenericChart.GenericChart
val NLRChart : GenericChart.GenericChart
val xHours : float []
val yCount : float []
val model' : Model
module Table

from FSharp.Stats.Fitting.NonLinearRegression
val LogisticFunctionAscending : Model
val initialParamGuess' : float []
module LevenbergMarquardtConstrained

from FSharp.Stats.Fitting.NonLinearRegression
val initialParam : xData:float [] -> yData:float [] -> cutoffPercentage:float -> float []
val solverOptions' : SolverOptions
val lineSolverOptions : initialParamGuess:float [] -> SolverOptions
val lowerBound : Vector<float>
val param : float
val abs : value:'T -> 'T (requires member Abs)
val upperBound : Vector<float>
val estParams : vector
val estimatedParams : model:Model -> solverOptions:SolverOptions -> lambdaInitial:float -> lambdaFactor:float -> lowerBound:vector -> upperBound:vector -> xData:float [] -> yData:float [] -> vector
val multipleSolverOptions : SolverOptions []
val initialParamsOverRange : xData:float [] -> yData:float [] -> steepnessRange:float [] -> float [] []
val solvO : SolverOptions
SolverOptions.InitialParamGuess: float []
val estimatedParamsWithRSS : model:Model -> solverOptions:SolverOptions -> lambdaInitial:float -> lambdaFactor:float -> lowerBound:vector -> upperBound:vector -> xData:float [] -> yData:float [] -> vector * float
val minBy : projection:('T -> 'U) -> array:'T [] -> 'T (requires comparison)
val snd : tuple:('T1 * 'T2) -> 'T2
val fst : tuple:('T1 * 'T2) -> 'T1
val fittingFunction' : (float -> float)
val fittedY : (float * float) []
val zip : array1:'T1 [] -> array2:'T2 [] -> ('T1 * 'T2) []
val fittedLogisticFunc : GenericChart.GenericChart
val xDataS : float []
val yDataS : float []
val data : (float * float) []
val knots : float []
val spline : lambda:float -> x:float -> float
val lambda : float
module Spline

from FSharp.Stats.Fitting
val smoothingSpline : data:(float * float) [] -> basispts:float [] -> (float -> float -> float)
val fit : lambda:float -> GenericChart.GenericChart
val sprintf : format:Printf.StringFormat<'T> -> 'T
val rawChartS : GenericChart.GenericChart
val smoothingSplines : GenericChart.GenericChart