java/src/liba/regress_simple.java

   1 package liba;
   2
   3 import java.nio.ByteBuffer;
   4
   5 /** simple linear regression */
   6 public class regress_simple {
   7     ByteBuffer ctx;
   8     static {
   9         System.loadLibrary("a");
  10         clinit();
  11     }
  12
  13     static final native void clinit();
  14
  15     final native void init(double coef, double bias);
  16
  17     /**
  18      * construct a new {@link regress_simple} object
  19      *
  20      * @param coef regression coefficient
  21      * @param bias intercept
  22      */
  23     public regress_simple(double coef, double bias) {
  24         init(coef, bias);
  25     }
  26
  27     /**
  28      * get regression coefficient for simple linear regression
  29      *
  30      * @return regression coefficient
  31      */
  32     public final native double coef();
  33
  34     /**
  35      * set regression coefficient for simple linear regression
  36      *
  37      * @param coef regression coefficient
  38      * @return {@link regress_simple}
  39      */
  40     public final native regress_simple set_coef(double coef);
  41
  42     /**
  43      * get intercept for simple linear regression
  44      *
  45      * @return intercept
  46      */
  47     public final native double bias();
  48
  49     /**
  50      * set intercept for simple linear regression
  51      *
  52      * @param bias intercept
  53      * @return {@link regress_simple}
  54      */
  55     public final native regress_simple set_bias(double bias);
  56
  57     /**
  58      * calculate predicted value for simple linear regression
  59      *
  60      * @param val independent variable
  61      * @return predicted value
  62      */
  63     public final native double eval(double val);
  64
  65     /**
  66      * calculate predicted value for simple linear regression
  67      *
  68      * @param val dependent variable
  69      * @return predicted value
  70      */
  71     public final native double evar(double val);
  72
  73     /**
  74      * calculate predicted value for simple linear regression
  75      *
  76      * @param x predictor data, specified as a numeric matrix
  77      * @param y response data, specified as a numeric vector
  78      * @param x_mean mean of predictor data
  79      * @param y_mean mean of response data
  80      * @return {@link regress_simple}
  81      */
  82     public final native regress_simple ols_(double[] x, double[] y, double x_mean, double y_mean);
  83
  84     /**
  85      * calculate predicted value for simple linear regression
  86      *
  87      * @param x predictor data, specified as a numeric matrix
  88      * @param y response data, specified as a numeric vector
  89      * @param x_mean mean of predictor data
  90      * @return {@link regress_simple}
  91      */
  92     public final native regress_simple olsx(double[] x, double[] y, double x_mean);
  93
  94     /**
  95      * calculate predicted value for simple linear regression
  96      *
  97      * @param x predictor data, specified as a numeric matrix
  98      * @param y response data, specified as a numeric vector
  99      * @param y_mean mean of response data
 100      * @return {@link regress_simple}
 101      */
 102     public final native regress_simple olsy(double[] x, double[] y, double y_mean);
 103
 104     /**
 105      * calculate predicted value for simple linear regression
 106      *
 107      * @param x predictor data, specified as a numeric matrix
 108      * @param y response data, specified as a numeric vector
 109      * @return {@link regress_simple}
 110      */
 111     public final native regress_simple ols(double[] x, double[] y);
 112
 113     /**
 114      * zeroing for simple linear regression
 115      *
 116      * @return {@link regress_simple}
 117      */
 118     public final native regress_simple zero();
 119 }