Видео

Without knowing your exact setup, it is difficult to say why. However, this error would appear if your sample is empty when running the regression. You can check step by step what happens in your data viewer to find out when exactly this occurs.

Hmmm. I think that winzoring don't have much to do with kurtosis and skewness assuming that you are only dealing with a few observations. Just my first thought though. Didn't check xD I would also assume that the same care should be taken with the dependent variable as with the independent ones.

@@SteffensClassroom After winsorizing the outliers in both the dependent and independent variables, the skewness and kurtosis results showed a normal distribution, producing the desired outcomes. However, I’m concerned that winsorizing the dependent variable could potentially distort the true results and introduce bias. For context, my sample consists of 110 companies, with firm performance measured over four consecutive years, resulting in a total of 440 observations.

@@mohitbalipa6013 Winzorizing is a data procedure which should be handled with great care. You can use it ofc, but you should document every step you take, and preferably show results with and without it. You can then show the effect of winzorizing to the reader, letting them draw their own conclusions.

No you cannot. Please consult the list and rules here: docs.google.com/document/d/1_YZU3WVeESPZ1tR4r9uVvFLpKcaJWBiU3GxJvIC-gtU/edit?usp=sharing You play 5 eeveelutions in total.

Ok, several things: 1. You literally didn't watch the entire video, and if you did, then you decided to shut off reason. 2. You didn't read the documentation. We provided literal paths to resources that explains why it's banned as a randomization practice. 3. I personally just attended the European finals in One Piece where someone in first round only pile shuffled as their means of randomization. You're either just being forcefully against the point of the video because you feel offended, or you're simply being naïve. We are not claiming that everybody who does it is being an intentional cheater. What we are saying, though, it's often a waste of time (depending on the goal) and that people ACTUALLY use it as a very 'innocent' practice to, in fact, cheat. You really have to try better bait than that..

A lot of players have their way of shuffling, that makes them feel that their deck is sufficiently random upon presenting it to your opponent. 7 riffles is in this case shown to achieve sufficiently randomization. Or rather, 8.something is, but 7 is inferred to be sufficient, as there is a sharp move towards randomness in terms of the total variantion distance.

That's simply not proper randomization of the entire deck. You use the words "declump". That means you are creating a deterministic order of your deck if you know the end result is cards won't be able to be anywhere within the deck. So per definition you're actually cheating according to the rules. You simply have to practice your shuffling better if you experience issues of the cards being in the same order as they were in the discard pile/trash/Graveyard from the most recent game. That's a simple fact and given. The math doesn't lie.

That's the same kind of randomness that some playlist shufflers use on streaming services. When you shuffle a music playlist, it's not really shuffled but arranged in a way such that it feels random, meaning songs that are close together in the playlist are spaced apart. But that's not randomness. Randomness by its very nature is going to sometimes cause clumps of the same cards.

Shuffling the individual piles together is just an extra step but I do think pile shuffling does do more than people are claiming. It assures the cards that are slightly warped get separated like the holos are always warped and end up together when riffling. All the arguments against pile shuffling assume you know the order of the deck and can track them. Nobody just pile shuffles I find the colt against it is really weird and petty

Hi! I'll try and see if I can find time :) *Steffen

Happy that you found it useful. Good luck with your paper!

Not use a Heckman (:

@@SteffensClassroom Can you suggest any alternatives?

I am not sure what you want to accomplish? You need to think about what the goal is. You could also simply transform your selection variable into a dummy? Again, I do not know what you wish to accomplish here.

I believe this is the same case as some other comments. You can check the answer there :)

Intentional draw but might as well be viewed as intentional defeat...

Feel free to mail me on the channel mail!

Hi! I believe that you are referring to the label option in the outreg2 command :)

Just once :)

This video is about how to standardize your variables. Other inquiries can be mailed, where a session can be booked.

I'll look into it

If you set your data as a panel via xtset (or tsset) then this is not a problem. It seems to me that you are trying to create the lags before telling Stata that it is a panel. Please lmk of setting it as a panel solves it.

@@SteffensClassroom Yes, you are right. I thought I had set it in the first place, but I hadn´t. Thank you for the help.

adjr2 You can see it in the help file. Type help outreg2.

Thank you for the comment! The collapse creates a new dataset on the basis of summary statistics of the old dataset. That is the purpose of the command. If you really want to keep the old dataset, simply wrap the collapse command in a preserve/restore group, creating a new dataset, saving it, and then merge it onto the old. I explained the preserve/restore and merge in two other videos.

Correct! Pokéball was too good in connection with the eeveelutions. In the original version of epic splash, Exodia pieces were used. They could not be searched with Pokéball as it was not a Pokémon. Since we now use stage 1 eeveelutions (since we have to use only Pokémon cards to be allowed at side events), Pokéball was simply too good, and was replaced with Dual Ball.

This is so cool! We need to organize a Worlds of some sort.

I turn one Lass'd and their entire hand was trainers

Hi! Thank you for the comment. It is difficult for me to see exactly what you did, but please note the use of ` ' around the references, such as when I write `var'. Feel free to mail me on the channel mail with the question. You could then attach the do-file you wrote.

Hi! Thank you for your comment. From what I understand, at 9:08, I mention indeed that there should not be too high of a correlation between IMR and the exclusion restrions. You mention that there should not be a significant correlation between your exclusion restriction and the dependent variable in the second stage. I am a bit confused here. Your exclusion restrictions are only there in the first stage, and not 'directly' in the second stage. Given what you show me here, (-0,1; p-value < 0,001). I wouldn't be too worried.

Math is scary :/

Did you make sure to install this command first? Ssc install outreg2?

Hi! Thank you for the comment. The way I like to see it is that since a hand that is a mulligan is a possible hand to draw, you still have to factor it in. However, with a mulligan, it would count as a reset. That is, each time you hit a mulligan, you get a new risk of getting a lone Lumineon V. We can in fact also try and simulate this. This is something I have been wanting to do for a long time, but simply didn't have the time to sit down and do. You can also look back at my answer on a different comment on this video :)

@@SteffensClassroom I see what you mean, yeah! It's kind like putting all the theory together, with one video we know What's the chance of starting with a bad pokemon What's the chance of mulligan What's the chance of having that hand, specifically including in the count hands that do not result in starting the game I think it's really interesting, especially considering now there are a couple decks that runs almost no basics, so adding one single support pokemon like lumineon could be really detrimental (a mulligan is probably almost better than starting with lone lumineon)

Yep :) That was exactly what I was trying to with this video, and the original version of this video (episode 2). I personally do not like to play Lumineon V, especially in the current meta as you typically run a lower amount of basics. It makes it very dangerous. This only becomes worse if you also have a deck where you can't attack or perhaps even remove Lumineon V from play.

@@SteffensClassroom yeah with my group we are having a lot of talks about the upcoming dragapult deck and we have discordant opinion about the rotom v. I, for one, don't like it, because as you say then you have to play a way to removed it (collapsed, turo, penny), and you kind of want to play the fss (which, while it gives you whatever card you need, it uses one of your 60 slots. What if rotom or the fss are prized? If the rotom is prized and you top deck fss you are not having a great time). And then, in the best case scenario, I use rotom once for 3 card (that can indeed be iono'ed or judged away), and now I have a 190hp 2 prize liability that I need another 2 to 4 deck slots to use properly... And I need more nest ball /ultra ball (in the case of lumineon) instead of maybe poffin... Idk, i think the advantages are smaller than the gains, especially adding the 'I might start with this and nothing else and I would have preferred a mulligan instead of starting lone rotom/lumineon'

I think you have to look at the potential upside versus the downside that you are describing. The situation you describe is similar (if not the same) as to what you see in Zard ex decks. Evidently, the choice is to play Rotom because the upside outweights the downside.

Hi! If they have a similar stub as for example 'feeling' in your example, then you can do this in a loop. The groups and loops video explains this.

@@SteffensClassroom Hi thanks again for the quick answer. If they are not the same - in my case ch22o_EN_1.0p and ch23p_EN_1.0p would there also be a loop for that?

The only time I encountered this, I simply went and changed the name of one of them (:

Yep, sad but true. Only give an average number. We can still have outliers, so you could miss a card even after 5k, 6k, 7k packs...

As explained. It is whether the two groups are significantly different from one another. That is, you need to check if the coefficient is significant or not.

So it more focused on a p and T scores rather than how large the coefficient is?

@@josephconway16 Hi again! In general yes, as you are trying to see if there are significant difference between the groups in question. There are other measures out there in order to test if a randomization is succesful (if two groups are significantly different from each other). You could go for a non parametric test such as a rank test, supposed you feel like the assumptions of a t-test is violated. You could also go for a resampling test.

Hi! Check one of the other comments. I believe I gave an answer there. In any case, I am unfortunately not certain :/