• Types of Statistics: 2 types – Descriptive and Inferential.
• Descriptive is based on numerical and graphical procedures. • Inferential draw conclusion based on sample. • Population is totality of units. • Results calculated from Population are called Parameters. • Parameters are denoted by Greek Letters { β, α, σ, μ…} • Population Mean is denoted by: μ • Parameters are Fixed Quantity. • Sample is representative part of Population. • Sample results are denoted by English Letters. • Sample mean is denoted by: X̅ • Sample results are called Statistic. • Statistic is a Variable Quantity. • Statistical Inference is a procedure used to draw conclusion about Population Parameter on the basis of Sample Statistic. • Quantitative (Measurable, Numerical) • Qualitative (non-numerical) • Qualitative is measured in – Nominal Scale (used for identification purpose), Ordinal Scale (used for ranking purpose). • Quantitative is measured in Interval Scale (show state), Ratio Scale(Height, weight). • Types of Data – Primary, Secondary. • Primary data is firsthand information. • Secondary data is organized form of data on which some stats is applied. • Comparative Analysis – measuring which is better among population. • Associative Analysis – finding relationship between two objects. • Predictive Analysis – Predicting future behavior on the basis of current behavior. • Measures of Central Tendency – Arithmetic Mean, Geometric Mean. Etc. • Measures of Dispersion (MOD) – Variation, Standard deviation, CV. Etc. • Types of MOD – Absolute (Range, Mean, SD, Variance etc.) Relative (CV, CR, CQ, Cm.) • Standard Deviation is square root of Variance. • Less Variance = More Reliable • High Average = More Reliable • Estimation Types – Point Estimation, Interval Estimation. • If probability is added in interval estimation it is called Confidence Interval. • Point estimate – single/fixed value. • Interval estimation – Define boundary value. • Interval estimation is more Secure than point estimation. • Six Sigma approach e.g., -6 +6 from SD. • Null Hypothesis always have sign of equality. • T-Test is used to test an assumption applicable to population. • Critical Region always depend on H1. • Alternate Hypothesis is also called Research Hypothesis. • While comparing two or more than two mean, T-pool or T-Prime test are used if samples are Independent. • If population variance is given & equal use t-pool. If not given, test variance is equal or not. • If population variance is unequal t-prime are applied. • If samples are Dependent use t-pair / Paired T test. • If S increase then tcal decrease. • If S decrease then Pvalue increase. • If n(sample size) decrease then tcal decrease. • If n increase then tcal increase. (they have direct relation) • Sample Size and Pvalue have Inverse Relation. • Types of Diagrams – Bar, Multiple Bar, Component, Pie. • Types of Graphs – Histogram, Historigram, Frequency Polygon, Frequency Curve, Ogive (cumulative frequency polygon) • Symmetrical Distribution – mean = median = mode. • In Minitab Ttab value = Pval • If Pvalue < α then Reject Ho. • If Pvalue > α then don’t Reject Ho. • Median divides dataset into 2 equal parts. • Quartile divides dataset into 4 equal parts. • Decile divides dataset into 10 equal parts. • Percentile divides dataset into 100 equal parts. • Median = Q2 • Median = D5 Median = P50 (contact for further details )