uiuc stats 100
Statistics 100 Spring 2017 Sections
Outline of Course Content
Experimental Design - Why randomized controls are key.
What the possible confounders in observational studies are.
Descriptive Statistics - mean, median, SD, histograms, normal curve, etc.
Linear Regression - correlation coefficient, regression equation, etc.
Probability
Statistics for Chance Numbers - expected value and Standard error of chance processes, probability histograms and convergence to normal curve. Focus is on developing simple chance models box models- drawing numbers at random from a box) that more complicated sampling processes can be translated into.
Sampling and Statistical Inference - Using sample means and percents to estimate population means and proportions, and attaching margins of errors to our estimates by computing confidence intervals. Why randomized sampling is key.
Hypothesis Tests-one sample and two sample Z-tests, t-tests and chi-square tests for goodness of fit and independence. Focus is on understanding how these tests depend on chance models.
Limits of Significance Tests- understanding what the P-value means and under what circumstances it is valid. (For example, hypotheses must be stated before looking at the data, the total number of experiments before significant results were found must be reported, etc.)
Read full...
Outline of Course Content
Experimental Design - Why randomized controls are key.
What the possible confounders in observational studies are.
Descriptive Statistics - mean, median, SD, histograms, normal curve, etc.
Linear Regression - correlation coefficient, regression equation, etc.
Probability
Statistics for Chance Numbers - expected value and Standard error of chance processes, probability histograms and convergence to normal curve. Focus is on developing simple chance models box models- drawing numbers at random from a box) that more complicated sampling processes can be translated into.
Sampling and Statistical Inference - Using sample means and percents to estimate population means and proportions, and attaching margins of errors to our estimates by computing confidence intervals. Why randomized sampling is key.
Hypothesis Tests-one sample and two sample Z-tests, t-tests and chi-square tests for goodness of fit and independence. Focus is on understanding how these tests depend on chance models.
Limits of Significance Tests- understanding what the P-value means and under what circumstances it is valid. (For example, hypotheses must be stated before looking at the data, the total number of experiments before significant results were found must be reported, etc.)
Read full...