Data science tools: Basic of Statistics

Identify the median by ordering data and locating the middle value, then calculate it for both raw data and grouped data using cumulative frequency and the median class.

Identify the mode as the most frequent value in a dataset, including unimodal, bimodal, and multimodal patterns, and learn to compute it for grouped data using class intervals.

Explore how the coefficient of variation expresses relative variability as a percentage, by dividing standard deviation by the mean to compare data with different units.

Explore the box plot, a five-number summary of min, max, Q1, median, and Q3, and learn to interpret and draw it with data in SPSS, R, Python, or Excel.

Identify how skewness shapes data distribution, including left negatively skewed, right positively skewed, and symmetric forms. Visualize with histograms, frequency polygons, and box plots, and relate mean, median, and mode.

Explore visualizing skewness via a histogram or polygon and quantify shape with Bowley's and Pearson's coefficients using quartiles, mean, median, and standard deviation, and identify left or right skewness.

Explore regression analyses, focusing on linear regression to model the relationship between x (predictor) and y (outcome), estimate alpha and beta with ordinary least squares, and predict y from x.

Explore the basic laws of probability, including the additive rule for unions and intersections and the complementary law. Apply conditional probability, independence, and the multiplication rule to problems.

Explore the binomial distribution, a two-outcome model with fixed, independent trials, and learn to compute probabilities of x successes using n, p, and the binomial formula.

Explore the Poisson distribution, where the probability of X occurrences in a given time interval or region depends on lambda, the mean, with the variance equal to lambda.

explains the normal distribution and its gaussian form, noting symmetry, mean median equality, and the empirical rule of 68%, 95%, and 99.7% within one, two, and three standard deviations.

Learn to work with the normal distribution by converting values to z-scores, and compute probabilities using the z-table and Excel. Apply these methods to examples like demand and semiconductor lifetimes.

Learn how the odds ratio compares the odds of an event between exposed and unexposed groups. A ratio greater than 1 indicates higher odds; less than 1 indicates lower odds.

Explore how screening tests are evaluated using sensitivity, specificity, and the gold standard. Build intuition with the confusion matrix and true/false positive and negative outcomes.

Define and apply sensitivity, specificity, and accuracy in screening tests. Identify true positives, true negatives, false positives, and false negatives, and explain predictive values and likelihood ratios.

Explore how sensitivity, specificity, and prevalence determine predictive values (positive and negative) using bayes' theorem for screening tests.