COST (GBP)

0.00

Maths › Approximation › Regression ›

Logistic

Evaluates the logistic regression curve built from a given set of points.

Controller: CodeCogs

Contents

Interface
Class Logistic
Logistic Once
Page Comments

Interface

C++

Excel

Class Logistic

The logit of a number $\inline p \in [0, 1]$ is

$\mathrm{logit}(p) = \mathrm{log}\left( \frac{p}{1 - p} \right) = \mathrm{log}(p) - \mathrm{log}(1 - p)$

(1)

The logit function is the inverse of the sigmoid, or logistic function. If $\inline p$ is a probability then $\inline p / (1 - p)$ is the corresponding odds, and the logit of the probability is the logarithm of the odds; similarly the difference between the logits of two probabilities is the logarithm of the odds-ratio, thus providing an additive mechanism for combining odds-ratios.

Logits are used for various purposes by statisticians. In particular there is the "logit model" of which the simplest sort is

$\mathrm{logit}(p_i) = a + b x_i$

(2)

where $\inline x_i$ is some quantity on which success or failure in the $\inline i$ -th in a sequence of Bernoulli trials may depend, and $\inline p_i$ is the probability of success in the $\inline i$ -th case. For example, $\inline x$ may be the age of a patient admitted to a hospital with a heart attack, and "success" may be the event that the patient dies before leaving the hospital (another instance of the reason why the words "success" and "failure" in speaking of Bernoulli trials should be taken with large doses of salt). Having observed the values of $\inline x$ in a sequence of cases and whether there was a "success" or a "failure" in each such case, a statistician will often estimate the values of the coefficients $\inline a$ and $\inline b$ by the method of maximum likelihood. The result can then be used to assess the probability of "success" in a subsequent case in which the value of $\inline x$ is known. Estimation and prediction by this method are called <em> logistic regression </em>.

As you may have noticed there is a link between the logistic and the linear regression methods, through the $\inline \mathrm{logit}$ function. In other words,

$\mathrm{logit}(p) = \mathrm{log}\left( \frac{p}{1 - p} \right) = a + b x$

(3)

Applying the exponential in both sides of the equality and doing further calculations, we arrive at the following relation

$p(x) = \frac{\mathrm{e}^{a + b x}}{1 + \mathrm{e}^{a + b x}} \qquad \mbox{or} \qquad p(x) = \frac{1}{1 + \mathrm{e}^{- a - b x}}$

(4)

where $\inline a$ and $\inline b$ are the parameters of the associated linear regression, intercept and slope.

Below you will find the regression graph for a set of arbitrary points, coloured in blue. The regression curve, displayed in red, has been calculated using this class.

Example 1

The following example evaluates the logistic curve for a given set of points, which is also displayed in the previous graph. The abscissas are equally spaced in the interval [10, 50] with a step of 5.

#include <codecogs/maths/regression/logistic.h>
#include <iostream>
#include <iomanip>
using namespace std;
 
int main() {
    double x[6] = {   28,    29,    30,    31,    32,    33};
    double y[6] = {.3333, .4000, .7778, .7778, .8000, .9333};
 
    Maths::Regression::Logistic A(6, x, y);
 
    cout << "Logistic regression values" << endl << endl;
    for (int i = 10; i <= 50; i += 5) {
        cout << "x = " << setw(3) << i << "  y = " << A.getValue(i);
        cout << endl;
  }
    return 0;
}

Output:

Logistic regression values
 
x =  10  y = 0.143118
x =  15  y = 0.233325
x =  20  y = 0.356719
x =  25  y = 0.502592
x =  30  y = 0.648024
x =  35  y = 0.770364
x =  40  y = 0.859406
x =  45  y = 0.917614
x =  50  y = 0.95304

References:

Wikipedia, http://en.wikipedia.org/wiki/Logistic_regression

Authors

Lucian Bentea (August 2005)

Source Code

Source code is available when you buy a Commercial licence.

Not a member, then Register with CodeCogs. Already a Member, then Login.

Members of Logistic

Logistic

Logistic(	int	`n`
	double*	`x`
	double*	`y`	)[constructor]

Initializes the class by calculating the slope and intercept of the corresponding linear regression function.

n	The number of points to consider
x	The x-coordinates of the n points
y	The y-coordinates of the n points

Logistic Once

doubleLogistic_once(	int	`n`
	double*	`x`
	double*	`y`
	double	`a`	)

This function implements the Logistic class for one off calculations, thereby avoid the need to instantiate the Logistic class yourself.

Example 2

The following graphs fits a single regression curve to the following values:

x = 23.2  y = 0.02
x = 33.3  y = 0.04
x = 33.5  y = 0.13
x = 34   y = 0.17
x = 34.2  y = 0.18
x = 34.2  y = 0.15
x = 34.4  y = 0.11

There is an error with your graph parameters for Logistic_once with options n=7 x="23.2 33.3 33.5 34 34.2 34.2 34.4" y="0.02 0.04 0.13 0.17 0.18 0.15 0.11" a=0:150 .input

Error Message:Function Logistic_once failed. Ensure that: Invalid C++

Parameters

n	The number of initial points in the arrays x and y
x	The x-coordinates for the initial points
y	The y-coordinates for the initial points
a	The x-coordinate for the output point

Returns

the interpolated y-coordinate that corresponds to a.

Source Code

Source code is available when you buy a Commercial licence.

Not a member, then Register with CodeCogs. Already a Member, then Login.

Last Modified: 14 Feb 08 @ 00:59 Page Rendered: 2022-03-14 17:44:37