C语言正态分布函数计算结果不符预期的技术求助

First off, let's break down the common pitfalls that lead to incorrect results when implementing the normal probability density function (PDF) with your constraints, then walk through a correct implementation and debugging steps.

Common Causes of Mismatched Results #

These are the most likely issues throwing off your calculations:

• Incorrect normalization constant: The normal PDF's denominator is sqrt(2 * π) * σ — it's easy to forget the 2 or use an imprecise value for π (stick to a double-precision value like 3.141592653589793).
• Sign error in the exponent: The exponent should be -0.5 * ((x - μ)/σ)^2 — omitting the negative sign or miscalculating the scaling factor will produce drastically wrong results.
• Order of operations mistakes: Failing to parenthesize (x - μ) before dividing by σ (e.g., writing x - μ/σ instead) will compute the wrong deviation from the mean.
• Accidental integer division: If μ or σ are integer types instead of double, dividing them can trigger integer division (e.g., 1/2 becomes 0 instead of 0.5).

Correct Implementation (Following Your Rules) #

Here's a compliant implementation that should match expected values:

#include <stdio.h>
#include <math.h>

double normal_pdf(double x, double mu, double sigma) {
    const double pi = 3.141592653589793; // Precise double-precision π
    double deviation = x - mu;
    double sigma_squared = sigma * sigma; // Use * instead of pow()
    double exponent = -0.5 * (deviation * deviation) / sigma_squared;
    double denominator = sqrt(2 * pi) * sigma;
    return exp(exponent) / denominator;
}

int main() {
    // Test case: Standard normal distribution (x=0, μ=0, σ=1)
    // Expected output: ~0.398942280401
    double result = normal_pdf(0.0, 0.0, 1.0);
    printf("Calculated PDF value: %.12f\n", result);
    return 0;
}

To compile correctly (don't forget linking the math library):

gcc your_filename.c -o normal_pdf -lm

Debugging Steps to Fix Your Code #

1. Validate the normalization constant: Compare your denominator to sqrt(2 * π) * σ. If your result is exactly double the expected value, you probably omitted the 2 in sqrt(2*pi).
2. Check the exponent calculation: For the standard normal test case above, the exponent should be 0 (so exp(0) = 1). Print intermediate values like deviation, sigma_squared, and exponent to spot discrepancies.
3. Ensure all variables are double: Even if you pass integer values (like 1), cast them to 1.0 to avoid integer division bugs.
4. Verify order of operations: Double-check that you're calculating (x - mu) first, then squaring it, then dividing by sigma*sigma.

Example Test Case Reference #

Expected output for standard normal at x=0: 0.398942280401
If your output is 0.797884560802, you missed the 2 in the denominator's sqrt(2*pi) term.
If your output is extremely small or large, you likely messed up the exponent's sign or scaling.

If you share your specific code snippet, we can pinpoint the exact issue even faster!

内容的提问来源于stack exchange，提问作者JJCFTW