void BartlettTest(TMtx *Data, int &NumDim, TVec *SignifVec, const double Alpha = 0.05);
Performs the Bartlett test for dimensionality of data.
| # | Name | Type | Description |
|---|---|---|---|
| 1 | Data | TMtx * | Data to be analyzed. |
| 2 | NumDim | int & | Returns the number of dimensions needed to explain the non random variation in data. If NumDim is less than zero, the test cannot be applied to this data. |
| 3 | SignifVec | TVec * | Returns calculated significance probabilities for one, two ... variables (dimensions). |
| 4 | Alpha = 0.05 | const double | Desired significance. |
Examples
#include "MtxExpr.h"
#include "Units.Statistics.h"
using namespace Dew;
using namespace Dew::Math;
using namespace Dew::Stats;
using namespace Dew::Stats::Units::Statistics;
void Example()
{
int numdim;
Matrix data(0,0);
Vector signifVec(0);
data.LoadFromFile("PCA_Data.mtx"); // load test data
BartlettTest(data, numdim, signifVec, 0.05); // Alpha = 5%
// returns numdim and signifVec
}See Also: Statistics::PCA
Declared in Dew::Stats::Units::Statistics · Dew.Stats/Units.Statistics.h · Cross-compiler