R-Fundamentals

This page i will show you the fundamental of R and in my project, convert a book to web version, every term will be explained and will be illustrated by videos.
Factor with two more levels
Recognize
Response variable
explanatory variables
explanatory variables continuous or categorical
Continuous measurement, a count, a proportion, a time-at-death or a category

Keys lead to the appropriate statistical method
1. Explanatory variables
- All explanatory variables continuous----> Regression
- All explanatory variables categorical-----> Analysis of Variance(Anova)
- Explanatory variables both continuous and categorical -----> Analysis of covariance
2. Response Variable
- Continuous                      Normal regression, Anova or Ancova
- Proportion                      Logistic regression
- Count                           Binary logistic Analysis
- Binary                                   
- Time-at-death
x<- c(2,3,4) it is just simple <- to assign the object
Everything varies:
When measuring the same thing many time we get different results
discriminating between variation that is scientifically interesting and variation of nature(reflects background heterogeneity) this is the whole page about.
Significance
A result was unlikely to have occurred by chance
Null hypothesis
unlikely: it occurs less then 5% of the time.
Good and bad hypothesis
A good hypothesis is a falsifiable hypothesis
adsense of evidence is not evidence of adsence
We reject the null hypothesis when our data show that the null hypothesis is sufficiently unlikely.
p Values

there are 2 kinds of mistakes in the interpretation of statistical models
                                                      Actual situation
Null hypothesis                      True                  False
 We       accept            Correct decision             Type II
              reject                      Type I               Correct decision

Statistical Modelling: we are looking for values of parameters in a model that is best fitted to the data ( dont touch to the data to fit the model, it is sacrosanct).
the model is fitted to the data,  i cant say " the data were fitted to the model."
the best model: least unexplained variation, subject to the constraint that all parameter in the model should be statistically significant.
We want the model to minimal because of the principle of parsimony and adequate
Maximum Likelihood
parameters in the model should afford the 'best fit of the model to the data'
Given data, given our choice of model, what values of the parameters of that model make data most likely(best, or maximum likelihood)

Experimental design

replication to increase reliability

randomization to reduce bias

the principle of parsimony

the power of a statistical test

controls

spotting pseudoreplication and knowing what to do about it

the difference between experimental and observational data (non-orthogonalitity)

the Principle of parsimony

Controls: no controls, no conclusions

Replication: it's the n's that justify the means, Why? do the same thing to the different individuals we are likely to get different responses.

the repeated measurements

- must be independent

-must not form part of a time series

- must not be grouped together in one place

- must be of an appropriate spatial scale 

I myself study this book, and this also my first project to turn a book into a blog that every contents in this book can be understood by whoever want to study this subject.

when i read a new book scientific book, a lot of new phases and new definitions this stops to check on internet to understand the terms
How many replicates
30, a sample of 30 or more is a big sample, but less than 30 is a small one., repeats may be expensive, replicates my be down to 5
We need to know the variance or response variable
Experiment and pilot studies are important, this indicates the variance between initial units.
Power
the power of the test is the probability of rejecting the null hypothesis when it is fall.
Dataframe