Overview of Statistical Methods

Review of statistical methods
Show Details

1. Survival Analysis
   Survival Analysis is used to model time to event data with the purpose of comparing survival function among different groups, the events often refer to death or failure.
   Key definitions include
   
   • survival function
   • hazard function (log normal, log logistic, weibull distribution...)
     
   • cumulative hazard function
   • censoring (left, right, interval)
   • truncation
   Key methods include
   
   • Kaplan-meier estimator (product-limit)
   • Nelson-Aalen estimator
     
   • Life-table estimator
     
   • Logrank test (Mantel-cox test)
   • AFT (Accelerated Failure Time model)
     
   • Cox Proportional Hazards Model (semiparametric model)
   Key software packages
   
   • SAS/PROC LIFEREG
   • SAS/PROC LIFETEST
   • SAS/PROC PHREG
     
   • R/survival (Surv, survfit, survdiff, survreg, coxph)
     
     
2. Logistic regression
   
   Logistic regression (Logit model) is used for prediction of probability of occurrence of an event by fitting data to a logistic curve. The response variable is categorical (binary, ordered, nominal).
   Key definitions include:
   
   • logistic function
   • Sigmoid function
   • Odds, Odds ratio, relative risk, risk difference
   Key methods include:
   
   • Probit model
   • Logit model
   • multinomial logit model
   • proportional odds model (cumulative logit model)
   • generalized linear model
   Key software packages:
   
   • SAS/PROC LOGISTIC
   • SAS/PROC PROBIT
     
   • SAS/PROC GENMOD
   • R/stats (glm)
   • R/nnet (multinom)
   • R/mlogit
   • R/MASS (polr)
     
     
     
3. Categorical Data Analysis
   
   Categorical Data Analysis
   Key definitions:
   
   • Categorical variable (binary, ordinal, nominal; dichotomous, multichotomous)
   • frequency (cell, marginal total)
   • proportion (cell, marginal total)
   • risk, risk difference, relative risk, odds, odds ratio
     
   • frequency table (one variable)
   • contingency table (two variables)
   • multidimensional contingency table (three or more variables)
   • hypergeometric distribution
     
   • binomial distribution
   • multinomial distribution
   • product multinomial distribution
   Key methods:
   
   • measurement of association
   • tests for trend
   • tests and measures of agreement (kappa coefficient)
     
   • Fisher's Exact Test
   • chi-square test
     
   • Cochran-Mantel-Haenszel (CMH) statistics
   • Log linear model
     
   Key software package:
   
   • SAS/TABULATE
     
   • SAS/PROC FREQ
   • SAS/PROC CATMOD
   • SAS/PROC GENMOD
   • SAS/PROC LOGISTIC
   • SAS/PROC PROBIT
   • SAS/PROC CORRESP
   • R/base (table, ftable)
     
   • R/stats (prop.test, chisq.test, fisher.test, mantelhaen.test, glm)
     
4. 
   

Charting with SAS

Charting with SAS
Show Details

There are many procedures that can be employed to make charts, curves, figures...

What knowledge does a master-level Biostatistician should have?

Biostatistician is a very professional position. In many situations, we have to ask ourselves: what ability does a master-level Biostatistician should have?
Show Details

Theoretical Knowledge

• Central Limit Theorem (CLT)
• Neyman Pearson Theorem
• Bayes Theorem
  
• Least-Squares methods
• Maximum likelihood methods
• Likelihood-ratio test
• Z-test
  
• Simple Linear Regression for linear relationship of dependent variable with another variable
  
• General Linear Model for linear relationship of dependent variable with multiple independent variables
• Generalized Linear Model for non-normal dependent response, poison distribution, binomial distribution, log normal distribution
• One sample t-test to test the mean of normal sample
  
• Sign test to test the median
• F-test to compare the variance of two normal samples
  
• Student's t-test to compare the mean for independent two normal samples with equal variance
  
• Welch's t-test to compare the mean for independent two normal samples with unequal variance
• one way ANOVA to compare means among three or more normal samples with equal variance
• two-way classification to evaluate effects of the two factors, main effects and interaction effects
• mixed model to evaluate the fixed effects of the mean and the variance of random effects or dependent normal samples among groups, repeated ANOVA, longitudinal data analysis, crossover design
  

Applied Knowledge using SAS

• Familiar with SAS Language Syntax, DATA step, PROC step, Global Statements, Loop, Conditional Statement
• Write reusable SAS MACRO
  
• Data inputs in DATA step with SAS/Base, PROC import, infile function
  
• Data manipulation with PROC SORT, array, merge, PROC SQL, data transformation (Log, Power, Box-Cox, variance-stabilization transformation)
• Data representation with SAS/GRAPH, in the format of tables, listing, and figures. PROC TABULATE, PROC GPLOT
  
• Statistical Analysis with SAS/STAT, PROC Means, PROC Univeriate, PROC FREQ, PROC GLM, PROC REG, PROC GENMOD, PROC mixed
• Model diagnosis (Gaphics)
  
• Model selection (AIC, BIC, Bayes factor, step-wise, forward, backward)
  
• Result delivery with ODS (Output Delivery System), ODS html, ODS pdf, ODS PROC REPORT
  

Overview of SAS system

In the Statistical point of view, SAS system is a software package used to reading, preprocessing, representation, statistical analysis, reporting of results of data. The functions include:
Show Details

• Reading data
• Preprocessing data
• Representing data
• Analyzing data
• Reporting results

1. Reading data
   The raw data in different formats can be read, stored into SAS session temporarily or permanently. The process is usually done in DATA steps. The following lists major packages or procedures that deal with the data input.
   
   • PROC IMPORT
   • DATA step
     
   • PROC SQL
     
2. Preprocession data
   After datasets are available to a SAS session, you need to preprocess the data in order to represent, summarize and analyze the data. Preprocessing refers to the process of creating new dataset, creating/modifying variables, merging multiple datasets, combining/deleting observations, cleaning datasets by all kinds of programs. The following lists major procedures that deal with the preprocessing of data.
   
   • PROC SQL
     
3. 
   
4. 
   

Overview of Experimental Design

Study note for Design of Experiments (DoE)
Show Details

Experimental Design or Design of Experiments (DoE)

1. Key concepts:
   
   • Effect size
   • Fixed effects/Random effects
     
   • Power
   • Sample size
   • Population
   • Bias/Error
     
   • Randomization
   • Experimental Unit
   • Blocking
   • Stratification
     
   • Replication
     
   • Blinding/Masking
   • Factors
   • Levels
   • Confounding factor
     
   • Treatment/Control, Placebo
   • Parallel design
   • Balanced design
     
2. Design:
   
   • Factorial design
   • Row-Column design
   • Crossover design
     
3. Models:
4. Steps for a good design
   
   1. Define the problem
   2. Determinate the primary/secondary endpoints or dependent variables
   3. Determinate the potential independent variables
   4. Determinate the levels and combination of each independent categorical variables
   5. Determinate the sample size based on the assumed effects, model, and power
   6. Design optimization
   7. Model specification
   8. Data collection
   9. Data clean-up
   10. Data verfication
       
5. Experimental Design in clinical trial
   
   • Phase-I design
   • Phase-II design
   • Phase-III design
   • Phase-IV design
     
6. 
   

How to add/place the adsense gadget on the left and right side

By default, the Adsense gadget can be placed on the sidebar or footer or header or between posts as shown below(click the image to zoom in).

How about placing the adsense gadget at the left/right side of the main page? as shown below (click the image to zoom in). By doing this, the effect will be like the one shown on my blog homepage.

To do this, you need to change the layout in the HTML editing mode.

<table><tr>
<td valign='top'>LEFT Adsense gadget</td>
<td>Body Layout</td>
<td valign='top' >RIGHT Adsense gadget</td>
</tr></table>

Where the body layout is the code between body tag of the original HTML template file, and the left/right adsense gadget code is like this

<div id='sidebar-wrapper'>
<b:section class='sidebar' id='sidebar1' preferred='yes'>
<b:widget id='AdSense1' locked='false' title='' type='AdSense'/>
</b:section>
</div>

and

<div id='sidebar-wrapper'>
<b:section class='sidebar' id='sidebar2' preferred='yes'>
<b:widget id='AdSense2' locked='false' title='' type='AdSense'/>
</b:section>
</div>

How to add a logo on your blog?

Do you want to add a logo on your blog, for example, the logo on my blog.
There are many ways to do this job, but here is the simple way to do this.
Instead of putting the logo and description in the header widget, I put the logo as well as description in a HTML/Javascript gadget. Below is the code for the gadget.

<table><tr>
<td><br/>
<img src="$LOGO_Link" align="middle"/>
</td>
<td valign="top" align="left">
<h1>$TITLE</h1>
$DESCRIPTION
</td>
</tr></table>

Clear the Description text in the header gadget, Replace the $LOGO_Link, $TITLE, $DESCRIPTION in the HTML/Javascript gadget and Rearrange your layout such that the gadget is just under the header gadget.