Hey everyone. I would like to use GetDist for visualizing the least-squares analysis of a chemistry experiment. The simplified version of what I want to do is this:

1. A lab assistant will mix two unknown amounts of either Chemical 1 with Chemical 2 *or* Chemical 1 with Chemical 3 in a container. ([math], [math], and [math] respectively)

2. I have an instrument that can detect both the total energy produced and the reaction rate of the process to 10% certainty. ([math] and [math] respectively) . IE... [math] Joules, etc...

4. I have a code that, if give it the mixtures [math] or [math], it will give me back the theoretical TotEng and RRate for those mixtures. IE calculates [math]. From these I believe I can construct a [math] for each situation.

5. I believe my [math] for the two situations are:

[*] [math]

and for the second situation

[*] [math]

My question is: how do I visualize this with Getdist? More specifically:

1. What does my chains .txt file format need to look like, for the first case?: [math]? And for the second case?: [math]? Or does it need to be: [math] and [math]?

2. If it turns out that [math] is inconsistent with the data for all possible fractions of these two chemicals, how do I plot that correctly? If my chains file is of the form [math], how does GetDist know about [math], and [math], to plot them? Or if it's [math], how does it know about [math], and [math]? Or do I make the chain file [math] so GetDist has access to all parameters? Or do I use the addDerived functionality somehow?

Sorry if this is confusing. I just want to use GetDist to distinguish between what factions of chemicals most likely produced the data and whether or not different chemical combinations could be consistent with the data. And I would like to plot those inconsistencies if possible and am just confused the best way to set up the chain files to make such plots. Thanks.

PS. Is Cobaya in some way suitable for this problem? Thanks.

## Using GetDist for general data

### Re: Using GetDist for general data

**Some followup questions1:**

1. Doing some reading, I think I want the file structure to be:

[math] where TotEng and RRate are derived parameters since [math], etc... Thus my .paramnames needs to be like:

Code: Select all

```
c1 C_1
c3 C_3
toteng* TotEng
rrate* RRate
```

2. Do the derived parameters always have to come after the non-derived? I don't see documentation explicitly say that, but all the examples are that way.

3. To plot confidence intervals don't you have to know the number of degrees of freedom of your system? I don't see where we input that information. Is this something the code can just infer from the chains?

Thanks.