Month: October 2021

Need: Feature engineering is one of the most important skills needed in data science and machine learning. It has a major influence on the performance of machine learning models and even the quality of insights derived during exploratory data analysis (EDA). What

Need: Feature selection, the process of finding and selecting the most useful features in a dataset, is a crucial step of the machine learning pipeline. Unnecessary features decrease training speed, decrease model interpretability, and, most importantly, decrease generalization performance on

Need  : As outliers impact the statistical analysis , once they detected using univariate or multivariate outlier detection methods , we need to treat them properly  before using them in the  Analysis. Approach:   The following are the some of the

Need: An outlier is a data point in a data set that is distant from all other observations. A data point that lies outside the overall distribution of the dataset. Outliers can drastically change the results of the data analysis

Need:   Before conducting any experiment, we need to select the appropriate and optimum sample size to make sure sample should be representative of the population and the finding are generalized to the population. Also, optimum sample size required To Allow

Need:  Most of the statistical/Machine Learning techniques requires the normality assumption for significance testing as well as for modeling, but in the real time scenario most of the times the data distribution doesn’t satisfy the normality conditions. Solution Transformation is

Need:    While choosing the suitable test statistics for hypothesis testing one of the key factors is data distribution. If the data distribution is normal distribution then we will apply the parametric tests such as t, Anova, f—etc., if data is

  Need:  When you identified missing data has a pattern with other variables (Missing at Random MAR) imputing the missing values with single value (Mean/Mode) imputation does not make any sense. As missing data has some kind relation with other

  Need: Missing data is a common and exciting problem in statistical analysis and machine learning hence we need to understand the cause of the missingness (pattern) and need to impute properly otherwise the random imputation might effect the solution

Objective Propensity score methods allow investigators to estimate causal treatment effects  for observational or nonrandomized data. Need: In most of the classical experiments either in healthcare or market research studies covariates (features) are balanced through randomization so that treated and