What type of analysis(s) should I use?

[#1]

I am doing a project on collecting data (prices, categories, size) from local competitors that sell alcohol online, E-commerce. I am trying to figure out a competitive pricing analysis to do on JMP, our company vs competitors. I have done so far on one competitor a multiple regression, paired t-test, residual plot analysis, ANOVA, and simple descriptive statistics. any recommendations I should use? I want to be able to standardize a process of my analysis's for all competitors. 

[#2]

Welcome to the community

Hmmmm...I think I understand you want to do some benchmarking of competitor pricing, but what do you want ultimately? A model that sets prices to beat, or match competitor pricing?  Ultimately you want to increase profits? I would think you also need to have some understanding of what the competitors are paying (e.g., inventory, storage, shipping/delivery).  Not sure how you get this?  

You should start by listing what questions you are trying to answer.  For example, What are the actual prices being charged by company x?  How consistent is this pricing?  What affects this pricing (e.g., promotions, holidays, loyalty)?  What comparisons do you want to make (e.g., brand, alcohol type, quantity discounts, delivery charges, speed of delivery)?

These questions should help you develop a list of what input variables and responses you want in your data table.  One of those columns will be competitor name.  No need to do this separately for each competitor.

You listed some statistical tools/techniques, but without context those are just a list of tools.  For example, what were you regressing on?  Why?  How much experience do you have in statistical method application?  

[#3]

To add a bit to @statman 's input...wrt to analysis you did NOT list any data visualization methods in your list of techniques/methods. Did you plot the data? If not...stop and go back and plot the data in a way that is consistent and will yield insight with the questions @statman has recommended formulating. Graph Builder, the Distribution platform are your starting points. At a minimum I recommend plotting all variables in a time series fashion if there is a time element to the data. Plot the continuous variables in a univariate and multivariate fashion as appropriate for the variables. You are looking for the following; 1. Where's the middle? 2. What's the shape? 3. How spread out is the data? 4. Is there anything odd or suspicious wrt to the data that might make subsequent analysis problematic? If you have categorical/nominal variables plot those in a mosaic fashion. Asking essentially the same questions above...but from a nominal variables point of view. One small example, suppose in your data set you have a nominal variable called competitor name AND in one row in your JMP data table it is spelled, "Acme Liquor" and in a different row it is spelled "acme liquor"...well JMP will treat those two rows as different values for that variable and you'll have to repair the rows to make them consistent in spelling and case orientation. If you are going to do some modeling work...I'd first utilize either the Fit Y by X platform or the Multivariate platform to examine correlations among the dependent and independent variables. All of this should help insure that you have data that is relevant, makes sense, and is 'clean' so subsequent analysis is most efficient and effective wrt to those questions @statman recommends you formulate.