Statistics Algorithms logopeterindia.net
Knowledge Hub

Statistics Algorithms

20 statistics algorithms and techniques every data and AI practitioner should know — statistical foundations, probability and relationships, statistical inference, and advanced data analysis and machine learning. Knowing when to use each one, its assumptions and how to read its results matters more than memorising formulas.

20 algorithms•4 categories•Updated November 13, 2026

Statistical Foundations

5 entries

Describe where data is centred and how spread out it is.

01MeanThe arithmetic average of a set of values; sensitive to outliers. 02MedianThe middle value when data is sorted; robust to outliers and skew. 03ModeThe most frequent value; the only average that works for categorical data. 04VarianceThe average squared distance from the mean, measuring how spread out data is. 05Standard DeviationThe square root of variance, in the same units as the data.

Probability & Relationships

5 entries

Quantify uncertainty and how variables move together.

06ProbabilityA number between 0 and 1 measuring how likely an event is. 07Bayes' TheoremUpdates the probability of a hypothesis when new evidence arrives; the basis of Bayesian methods and Naive Bayes. 08CorrelationMeasures the strength and direction of a relationship between two variables; it does not prove causation. 09RegressionModels how a target variable depends on one or more predictors, for explanation and prediction. 10Normal DistributionThe bell-shaped distribution defined by its mean and standard deviation, central to many tests and models.

Statistical Inference

5 entries

Draw conclusions about a population from a sample.

11Hypothesis TestingTests a claim about a population by asking whether sample data is consistent with a null hypothesis. 12p-ValueThe probability of data at least this extreme if the null hypothesis were true; it is not the probability the hypothesis is true. 13Confidence IntervalA range of plausible values for a population parameter, built so that it covers the true value at a stated rate. 14T-TestCompares means, such as two groups or a sample against a reference value, when the population variance is unknown. 15ANOVACompares the means of three or more groups by splitting variance between and within groups.

Advanced Data Analysis & ML

5 entries

Tests and learning algorithms built on statistical ideas.

16Chi-Square TestTests whether observed counts in categories differ from expected counts, or whether two categorical variables are independent. 17Decision TreesSplits data with a sequence of feature-based rules to predict a class or a value; the building block of random forests and boosting. 18K-Means ClusteringGroups data into k clusters by assigning points to the nearest centroid and updating centroids. 19Principal Component Analysis (PCA)Reduces dimensionality by projecting data onto the directions of greatest variance. 20Time Series ForecastingPredicts future values from time-ordered data using methods such as exponential smoothing and ARIMA.

Source

1 entry

The post this page is based on.

2120 Statistics Algorithms Every Data & AI Enthusiast Should Know (Akhil Meesala, LinkedIn)LinkedIn post listing the 20 techniques in four groups, with the advice to learn each method's assumptions and how to interpret its results.

Related Pages

5 entries

Related research directories on this site.

22StatisticsStatistics hub. 23Machine Learning AlgorithmsAlgorithms used in machine learning. 24Statistics for Deep LearningStatistics concepts for deep learning. 25Probability Theory for AIProbability foundations for AI. 26Data SciencePlatforms, notebooks and data wrangling for data scientists.

Frequently Asked Questions

Why learn statistics before machine learning?

Statistics explains how data behaves, how much to trust a result and how to compare models fairly. Concepts such as distributions, variance, hypothesis tests and regression sit underneath most machine learning methods.

What is the difference between mean, median and mode?

The mean is the average, the median is the middle value and the mode is the most frequent value. On skewed data or data with outliers, the median usually describes the typical value better than the mean.

What does a p-value actually mean?

It is the probability of seeing data at least as extreme as yours if the null hypothesis were true. It does not give the probability that the hypothesis is true, and a small p-value does not by itself mean an effect is large or important.

When should I use a t-test, ANOVA or chi-square test?

Use a t-test to compare the means of two groups, ANOVA to compare the means of three or more groups, and a chi-square test to compare counts or test independence between categorical variables. Each has assumptions to check first.

Are decision trees, K-means and PCA really statistics?

They sit between statistics and machine learning. Decision trees and K-means learn patterns from data, and PCA is a statistical technique based on variance and covariance, so all three are commonly taught as part of the statistical toolkit for data work.