#1
Hi there,

I would like to ask a couple of basic questions:

1. What is the difference between estimation and hypothesis testing? I mean are they two options for inference? or estimation of the parameters comes first and then we apply the hypothesis testing?

2. How can I determine what is the most suitable statistical model for my dataset?

Please help me with these issues
Thank you in advance.
 
#2
Hi,

(1): The usual (frequentist) way is: You have a hypothesis, and you express this hypothesis in terms of a model. Let's say your hypothesis is that a plant grows if you give the plant more nutrients. Your model could be plant_length ~ alpha*amount_nutrients. In a second step, you fit this model to data, i.e. you estimate your parameters. And depending on the estimated value of alpha (and it's p-value) it is either likely or unlikely that your hypothesis is true

(2): You can compare differnt models (e.g. with different set of predictors) e.g. via the AIC-value