Accord.Statistics.Testing.Power.ZTestPowerAnalysis.ComputeEffect C# (CSharp) Method

ComputeEffect() public method

Computes the minimum detectable effect size for the test considering the power given in IPowerAnalysis.Power, the number of samples in IPowerAnalysis.Samples and the significance level IPowerAnalysis.Size.
public ComputeEffect ( ) : void
return void
        public override void ComputeEffect()
        {
            double Za = ZTest.PValueToStatistic(Size, Tail);
            double Zb = NormalDistribution.Standard
                    .InverseDistributionFunction(Power);

            double n = Math.Sqrt(Samples);
            double d = (Za + Zb) / n;

            Effect = d;
        }

Usage Example

Exemplo n.º 1
0
        /// <summary>
        ///   Estimates the number of samples necessary to attain the
        ///   required power level for the given effect size.
        /// </summary>
        ///
        /// <param name="sampleSize">The number of observations in the sample.</param>
        /// <param name="power">The desired power level. Default is 0.8.</param>
        /// <param name="alpha">The desired significance level. Default is 0.05.</param>
        /// <param name="hypothesis">The alternative hypothesis (research hypothesis) to be tested.</param>
        ///
        /// <returns>The required number of samples.</returns>
        ///
        public static ZTestPowerAnalysis GetEffectSize(int sampleSize, double power   = 0.8, double alpha = 0.05,
                                                       OneSampleHypothesis hypothesis = OneSampleHypothesis.ValueIsDifferentFromHypothesis)
        {
            var analysis = new ZTestPowerAnalysis(hypothesis)
            {
                Samples = sampleSize,
                Size    = alpha,
                Power   = power,
            };

            analysis.ComputeEffect();

            return(analysis);
        }
All Usage Examples Of Accord.Statistics.Testing.Power.ZTestPowerAnalysis::ComputeEffect