This Statistics course aims to address the growing need for a practical and systematic introduction to statistics at work.
Course overview:
It is designed to give delegates a strong foundation in classical statistics,both descriptive and inferential,that will allow them not only to accurately interpret statistical results but also equip them with the requisite knowledge to design and carry out powerful statistical tests. The foundations will be laid to enable delegates to characterize the data they have,decide on the best summary and visual representations and,most appropriate tests to carry out. These are vital skills in the age of data. Content in General With the statistical content we cover in this course,delegates will acquire a general framework that will enable them to decide-on and perform a relatively wide range of common statistical procedures and tests on data. We will start by identifying different categories of data and the various descriptive statistical measures we can perform on them. This will be followed by inferential statistics which will include population and sampling distributions,confidence intervals and hypothesis testing. We will also cover concepts of probability,including distributions,on a just-in-time basis as required by our principal focus,statistical and inference. Approach: We believe in learning-by-doing,so we have taken a problem-solving approach to delivering our training. The course is broken into sessions,each centred on a few related core concepts and skills. The relevant theory is discussed at the beginning of the session,in a just-in-time approach. This is followed by an illustrative example. For the second half,which will generally be most of the session,the delegates are expected to solve relevant problems of graduated difficulty. Immediate practice helps delegates consolidate their understanding of concepts on which we build gradually. Example solutions will be available for the delegates to take away at the end of the course. We use the statistical language R and Python (with various relevant packages). So,delegates are expected to be proficient in one of the programming languages.
Course objectives:
This course aims to provide the delegate with the knowledge to be able to: • Calculate basic descriptive statistical measures such as • Mean,Median,Mode • Variance,Standard Deviation and Quartiles • Understand the framework of classical statistical inference • Understand and use discrete and continuous probability distributions including • Binomial,Bernoulli,Poisson,Normal,t-Distribution,Chi-Square,F-Distribution • Understand and/or calculate • Sampling bias and strategies • Population and sample distributions • Point estimates and confidence intervals • Design,choose and perform a hypothesis test • Understand and perform basic linear regression • Produce various visual representation (or plots) of data
Course content:
Statistics Training Course Course Contents – DAY 1: DESCRIPTIVE STATISTICS Course Introduction • Administration and Course Materials • Course Structure and Agenda • Delegate and Trainer Introductions Session 1: INTRODUCTION • What is Statistics and Why? • Statistical traditions: Classical and Bayes • Data types and classification Session 2: SUMMARY STATISTICS FOR CATEGORICAL DATA • Absolute Frequency (Count) • Relative Frequency • Grouped Frequency • Cumulative Frequency Session 3: SUMMARY STATISTICS FOR NUMERICAL DATA; LOCATION • Measures of central tendency • Mean • Median • Mode Session 4: MEASURES OF DISPERSION AND GRAPHICAL REPRESENTATION • Measures of dispersion • Range (Full,Interquartile etc.) • Variance and Standard Deviation • Graphical representation • Univariate • Discrete data,Frequency plots,Histograms • Continuous data,dotplot,stem-and-leaf plot,histograms,boxplots • Bivariate • scatter plot • line plot Statistics Training Course Course Contents – DAY 2: THE FRAMEWORK FOR STATISTICAL INFERENCE Session 5: DISCRETE NUMERICAL VARIABLE • Example of coin toss • Is my coin biased? • How confident am I? (Binomial distribution) • Discrete probability distributions (PMF) • Bernoulli • Binomial • Poisson Session 6: CONTINUOUS NUMERICAL VARIABLE: POPULATION MEAN • Terminology: Population vs Sample • Sample distribution • Mean value theorem and Normal distribution • Point estimate of mean • Standard error and confidence interval Session 7: SAMPLING CONSIDERATIONS • Sampling bias • Sampling strategies • Sampling statistics distribution Statistics Training Course Course Contents – DAY 3: HYPOTHESIS TESTING FOR NUMERICAL DATA Session 8: CONTINUOUS PROBABILITY DISTRIBUTIONS • Normal and Student-T distributions • Chi-Squared Distribution • F-Distribution Session 9: HYPOTHESIS TESTING FOR NUMERICAL DATA • Single population • Test against a hypothesized mean (Z / T – Test) • Type-I and Type-II errors • Test against a hypothesized variance (Chi-Squared Test) • Two populations • Test a difference in mean • Power calculations • Paired data • Comparison of variances • ANOVA Statistics Training Course Course Contents – DAY 4: HYPOTHESIS TESTING FOR CATEGORICAL DATA AND REGRESSION Session 10: HYPOTHESIS TESTING FOR CATEGORICAL DATA • Test for a proportion • Difference of proportions • Goodness of Fit • Test for independence Session 11: REGRESSION; BIVARIATE DATA • Correlation • Linear regression
TARGET AUDIENCE:
Who will the Course Benefit? This course will benefit anyone who works or intends to work with data. This course also offers a solid statistical foundation to professional or aspiring Data Scientists.
Prerequisites:
Requirements Mathematics: Ideally A-Levels but at a minimum GCSE level is required,Delegates will be expected to understand simple formulae,percentages,proportions and limits,and interpret simple formulae and graphs. They will also be expected to perform basic algebraic manipulations,add and subtract fractions. Programming: Delegates must also have an ability to program in Python or R up to and including defining and using functions. Our Introduction to Programming (in Python) course or our R Programming course will cover all programming requirements for this course.