[lecture_comparison][FiniteMarkovChains]equation reference numbering issue

[sayaikegawa]

Hi @mmcky ,

I found that the number of equation reference in MYST does not coincide with that in RST in lecture 12: Finite Markov Chains.

As the screen shot shows, the equation which should be referenced as (12.4) is referenced as (12.5) in MYST. And there is not an equation which is referenced as (12.4) in MYST. ( This means the equation which has a label before (12.5) is (12.3)) (left: RST, right:MYST)

I compared both files in VS code, but nothing seems to be wrong in MYST.

So far, I found this issue only in this lecture 12. So I have no idea why this happens. I'll comment when I find this issue in other lecture.

[mmcky]

thanks @sayaikegawa -- so the text reference is still ok but the numbering is a little different. Is that right?

Does that lecture have a 12.4?

[sayaikegawa]

Hi @mmcky , yes, that's right.

The lecture does not have 12.4. I have no idea why 12.4 is skipped. And VS code seems ok...

[mmcky]

that is strange -- nice pickup @sayaikegawa

[mmcky]

@AakashGfude could you take the lead on this issue re: numbering. Might be part of sphinx-multitoc-numbering?

[AakashGfude]

@mmcky @sayaikegawa this block of maths is causing the issue:

(mc_fdd)=
$$
\psi_{t+1}(y) = \sum_{x \in S} P(x,y) \psi_t(x)
$$

because it has a label, it assumes itself as 12.4, but since it is not a ```{math} myst directive. But, it does not display it in HTML?

[AakashGfude]

I think a reasonable solution would be to use ```{math} instead here for that block?

[sayaikegawa]

Hi @AakashGfude ,

In MyST document, it says dollar delimited math can have label as follows:

So, I wonder whether the equation has a label or not. Do you mean (mc_fdd)= is a label?

I'm not sure well but how about getting rid of (mc_fdd)= & (mc_fddv)=?

(mc_fdd)=
$$
\psi_{t+1}(y) = \sum_{x \in S} P(x,y) \psi_t(x)
$$

There are $n$ such equations, one for each $y \in S$.

If we think of $\psi_{t+1}$ and $\psi_t$ as *row vectors* (as is traditional in this literature), these $n$ equations are summarized by the matrix expression

(mc_fddv)=
```{math}
:label: fin_mc_fr

\psi_{t+1} = \psi_t P

[AakashGfude]

@sayaikegawa just tested it out and the MyST document format works nicely for numbering the equations with $$.
which would turn this equation to:

$$
\psi_{t+1}(y) = \sum_{x \in S} P(x,y) \psi_t(x)
$$(eqn:mc_fdd)

This then becomes (12.4)

Else, if we don't want eqn nos. then we can get rid of (mc_fdd)= & (mc_fddv)=, if that's what we prefer.

[sayaikegawa]

@AakashGfude , I see. Thank you.
I think if we follow the numbering in the current website RST,
the equation

(mc_fddv)=
```{math}
:label: fin_mc_fr

\psi_{t+1} = \psi_t P```

should be (12.4) instead of

$$
\psi_{t+1}(y) = \sum_{x \in S} P(x,y) \psi_t(x)
$$

[mmcky]

thanks @sayaikegawa and @AakashGfude it would be good to match the current live site for numbering (assuming that is correct).

[mmcky]

@sayaikegawa @AakashGfude I suspect this is due to the mix of link targets that are being specified. I have opened QuantEcon/lecture-python#367 which will hopefully fix this issue by removing some link targets that are unused.

[mmcky]

thanks @sayaikegawa this is now fixed by: QuantEcon/lecture-python#367

https://60249540dd68788589e83402--nostalgic-wright-5fa355.netlify.app/finite_markov.html#equation-fin-mc-fr