Skip to content
Calcrivo

Likelihood Calculator

Calculate the likelihood of observed data given a set of model parameters.

Inputs

Comma-separated data points

Log-Likelihood

-2.7090

Likelihood

0.06660626

Observations

5

Step by step

  1. Number of observations

    n

    = 5

  2. Log-likelihood (Gaussian): Σ[-0.5ln(2πσ²) − (x−μ)²/(2σ²)]

    n=5, μ=5, σ=0.5

    = -2.708957

  3. Likelihood: exp(log-likelihood)

    e^(-2.7090)

    = 6.6606e-2

How it works

The likelihood function measures how probable the observed data is under a given set of model parameters. For Gaussian-distributed data, the log-likelihood is the sum of log-probabilities of each observation under the assumed normal distribution N(μ, σ²). Maximum Likelihood Estimation (MLE) finds parameters that maximize this value.

Formula

Gaussian Log-Likelihood

LL = sum(-0.5*ln(2*pi*sigma^2) - (x_i - mu)^2 / (2*sigma^2))

mu
Assumed population mean
sigma
Assumed standard deviation

Frequently Asked Questions

Why use log-likelihood instead of likelihood?

Multiplying many small probabilities leads to numerical underflow. Taking logs converts products to sums, which is numerically stable and mathematically equivalent for optimization (maximizing log-likelihood maximizes likelihood).

How does this relate to training neural networks?

Training with cross-entropy loss is equivalent to maximizing the log-likelihood of the training data under the model's predicted distribution — MLE is the statistical foundation of most supervised learning.

You might also need