0.5
1.648721
0.622459
62.25
%
1
1
0.122459
0.5
1.648721
0.622459
62.25
%
1
1
0.122459
The Logistic Regression Calculator computes the predicted probability, odds, and log-odds for a binary outcome using a single-predictor logistic regression model. Logistic regression is one of the most widely used statistical methods in the social sciences, medicine, epidemiology, marketing, and machine learning for modeling binary dependent variables — outcomes that take only two possible values such as pass/fail, yes/no, disease/no disease, or purchase/no purchase.
Unlike ordinary linear regression, which predicts a continuous numerical outcome, logistic regression models the probability that the dependent variable equals one of two categories. The model achieves this through the logistic (sigmoid) function, which maps any real-valued linear predictor onto the interval (0, 1), making it interpretable as a probability. This elegant mathematical transformation ensures that predicted probabilities never fall below 0 or exceed 1, a problem that plagues linear probability models.
The parameters of a logistic regression model — the intercept (b0) and the slope coefficient (b1) — are typically estimated using maximum likelihood estimation (MLE) from observed data. Once these parameters are known, you can use this calculator to compute the predicted probability for any given value of the predictor variable. This is especially useful in clinical decision-making, where a physician might want to know a patient's probability of developing a condition given their age, blood pressure, or biomarker level.
The three outputs of this calculator provide complementary perspectives on the same prediction. The log-odds (also called the logit) is the raw linear predictor, which can range from negative to positive infinity. The odds represent the ratio of the probability of the event occurring to the probability of it not occurring. The probability gives the intuitive 0-to-1 likelihood of the event. Understanding all three representations is essential for correctly interpreting logistic regression results in published research and professional practice.
Logistic regression has become a cornerstone of predictive analytics and machine learning. It serves as the building block for more complex models including multinomial logistic regression (for multi-class outcomes), neural networks (where the sigmoid function acts as an activation function), and ensemble methods. Despite its simplicity, logistic regression often performs remarkably well in practice and remains the first-choice model for many binary classification problems due to its interpretability, computational efficiency, and strong theoretical foundation.
The logistic regression model relates a predictor variable x to the probability of a binary outcome through the logistic (sigmoid) function. The model has two parameters: the intercept b0 and the slope coefficient b1.
First, the calculator computes the linear predictor (log-odds): $$\text{logit}(p) = b_0 + b_1 \cdot x$$
This value can range from negative infinity to positive infinity. It represents the logarithm of the odds of the event occurring.
Next, the linear predictor is transformed into a probability using the logistic function: $$p = \frac{1}{1 + e^{-(b_0 + b_1 x)}}$$
The logistic function is an S-shaped (sigmoid) curve that maps any real number to the interval (0, 1). When the linear predictor equals zero, the probability is exactly 0.5. As the linear predictor increases toward positive infinity, the probability approaches 1; as it decreases toward negative infinity, the probability approaches 0.
Finally, the odds are computed as the ratio of the probability to its complement: $$\text{odds} = \frac{p}{1 - p}$$
Odds greater than 1 indicate the event is more likely than not; odds less than 1 indicate the event is less likely than not. The coefficient b1 has a direct interpretation in terms of odds: a one-unit increase in x multiplies the odds by e^b1 (the odds ratio).
The log-odds value is the raw linear predictor. A positive log-odds means the event is more likely than not (probability > 0.5), while a negative log-odds means it is less likely. A log-odds of zero corresponds to a probability of exactly 0.5 (50-50 chance). The odds value tells you how many times more likely the event is to occur than not. For instance, odds of 3.0 mean the event is three times as likely to happen as not happen. The probability is the most intuitive output, giving the estimated chance (between 0 and 1) that the event occurs for the given predictor value. A probability above 0.5 is typically classified as a positive prediction in binary classification tasks.
Inputs
Results
With intercept -3.5, age coefficient 0.08, and age 55: log-odds = -3.5 + 0.08*55 = 0.9, giving probability ~71.1%. This patient has a relatively high predicted risk.
Inputs
Results
With intercept -1.2, coefficient 0.3, and predictor value 4: log-odds = -1.2 + 0.3*4 = 0.0, yielding probability = 0.5. This is the decision boundary where the model is equally uncertain.
The logistic function $$f(z) = \frac{1}{1+e^{-z}}$$ is a sigmoid curve that maps any real number to the interval (0, 1). It is used in logistic regression because it ensures predicted probabilities remain between 0 and 1, unlike linear regression which can produce impossible probability values below 0 or above 1. The function also has convenient mathematical properties that make maximum likelihood estimation tractable.
The coefficient b1 represents the change in log-odds for a one-unit increase in the predictor variable x. To convert this to an odds ratio, compute e^b1. For example, if b1 = 0.5, the odds ratio is e^0.5 = 1.649, meaning the odds of the event increase by about 64.9% for each one-unit increase in x. This multiplicative interpretation makes odds ratios the standard way to report logistic regression effects.
Probability is a value between 0 and 1 representing the likelihood of an event. Odds are the ratio of the probability of the event to the probability of its complement: odds = p/(1-p). A probability of 0.75 corresponds to odds of 3 (the event is 3 times more likely than not). While probability is bounded between 0 and 1, odds range from 0 to infinity, and log-odds range from negative infinity to positive infinity.
Yes, the full logistic regression model supports multiple predictors: $$\text{logit}(p) = b_0 + b_1 x_1 + b_2 x_2 + \ldots + b_k x_k$$. This calculator implements the simplified single-predictor version for educational purposes and quick calculations. Each coefficient in the multiple predictor model has the same interpretation: the change in log-odds for a one-unit increase in that predictor, holding all other predictors constant.
A predicted probability of 0.5 represents the decision boundary in binary classification. The model is equally uncertain about both outcomes. This occurs when the linear predictor (log-odds) equals zero, i.e., when b0 + b1*x = 0. The predictor value at this threshold is x = -b0/b1. Values of x above this threshold yield probabilities above 0.5 (if b1 > 0), and values below yield probabilities below 0.5.
Logistic regression coefficients are estimated using maximum likelihood estimation (MLE). Unlike linear regression which has a closed-form solution (OLS), logistic regression requires iterative optimization algorithms such as Newton-Raphson or iteratively reweighted least squares (IRLS). Software packages like R, Python (scikit-learn, statsmodels), SAS, and SPSS perform this estimation automatically. This calculator assumes you already have estimated coefficients.
Roboculator Team
The Roboculator Team explains calculations, planning tools, and practical formulas in clear language for real-life situations.
How helpful was this calculator?
Be the first to rate!
Linear Regression Calculator
Regression & Correlation Analysis
Simple Linear Regression Calculator
Regression & Correlation Analysis
Multiple Regression Calculator
Regression & Correlation Analysis
Polynomial Regression Calculator
Regression & Correlation Analysis
Exponential Regression Calculator
Regression & Correlation Analysis
Logarithmic Regression Calculator
Regression & Correlation Analysis
