Linear Regression Model.Get Prediction Bandwidth Method
Definition
Namespace: Numerics.NET.Statistics
Assembly: Numerics.NET (in Numerics.NET.dll) Version: 9.0.3
Assembly: Numerics.NET (in Numerics.NET.dll) Version: 9.0.3
Overload List
Get | Gets the width of the prediction band around the best-fit curve at the specified point. |
Get | Gets the width of the prediction band around the best-fit curve at the specified point. |
GetPredictionBandwidth(Vector<Double>)
Gets the width of the prediction band around the best-fit curve at the specified point.
public double GetPredictionBandwidth(
Vector<double> x
)
Parameters
Return Value
DoubleThe confidence interval for the prediction at x.
Remarks
This method returns the distance from the predicted value to the upper and lower bound of the confidence interval for a new observation at x. This interval is wider than the interval returned by GetConfidenceBandwidth(Vector<Double>), which is the interval for the value predicted by the model.
GetPredictionBandwidth(Vector<Double>, Double)
Gets the width of the prediction band around the best-fit curve at the specified point.
public double GetPredictionBandwidth(
Vector<double> x,
double confidenceLevel
)
Parameters
- x Vector<Double>
- The point at which to evaluate the bandwidth.
- confidenceLevel Double
- The confidence level of the prediction band.
Return Value
DoubleThe confidence interval for the prediction at x.
Remarks
This method returns a confidence interval for a new observation at x. This interval is wider than the interval returned by GetConfidenceBandwidth(Vector<Double>, Double), which is the interval for the value predicted by the model.