Index | Recent Threads | Who's Online | Search

Posts: 90    Pages: 9    Prev 1 2 3 4 5 6 7 8 9
Last Post
This topic has been viewed 3025 times and has 89 replies
yuvaltaylor
Re: New engine released

If you're using quarterly numbers for companies that report semiannually, you might want to make a few minor adjustments. Quarterly numbers from the income statement and cash flow statement will be twice as high as those for other companies (TTM, annual, and balance-sheet-item numbers will be fine). Almost all fundamental items will give you NA for odd-numbered quarters and formulas that end in PQ.

Does this apply to both engines or just the Factset engine?

I was discussing the Compustat engines (both of them). The FactSet engine hasn't been changed yet, so you won't get NAs, but you'll get the incorrect period along with double the quarterly values for income-statement and cash-flow-statement items.

Yuval Taylor
Product Manager, Portfolio123
invest(igations)
Any opinions or recommendations in this message are not opinions or recommendations of Portfolio123 Securities LLC.

May 25, 2020 4:46:02 PM       
yuvaltaylor
Re: New engine released


Q. Is this how semiannuals will be treated in FactSet going forward?


We haven't come up with a permanent solution yet. We'll be changing the FactSet engine to operate like the Legacy engine shortly, but after that we are hoping to come up with a mechanism that will alleviate these problems. And we will be exposing semiannual companies as such so that elaborate workarounds won't be necessary.

Yuval Taylor
Product Manager, Portfolio123
invest(igations)
Any opinions or recommendations in this message are not opinions or recommendations of Portfolio123 Securities LLC.

May 25, 2020 4:48:11 PM       
yuvaltaylor
Re: New engine released


I would urge caution about relying on the share count from the financial statements. It can be stale when there has been a large share issuance or buyback after the closing date of the most recent financial statements. I have been burned on this in the past. Caveat emptor. You don't want to be trading based on stale information when the rest of the world knows up to date information.

SharesCur(0) has the advantage in that it is up to date as of the latest trading day. OTOH, as you brought up, SharesCur(0) is not accurate when there are multiple share classes and such as in the examples you gave above.

Solution: We need a up to date share count that includes all share classes.


You make some excellent points here, Chaim. Here's what I use:

eval(splitcount(100)=0,mktcap/price + isna(sharescur(0) - sharescur(5*max(0,weeksintoq)),0) + max(0,sharesfdq-sharesq),sharesfdq)

It's not perfect. If you can suggest a better formula, I'd be glad to consider it.

There is some merit in not looking at sharescur(0). After all, nothing else you're using when looking at a company's fundamentals reflects post-filing changes. So if a company buys back some shares four weeks after its filing and you use sharescur(0), your P/E will change but your ROA and profit margin won't.

Yuval Taylor
Product Manager, Portfolio123
invest(igations)
Any opinions or recommendations in this message are not opinions or recommendations of Portfolio123 Securities LLC.

May 25, 2020 4:58:53 PM       
Chipper6
Re: New engine released

There is some merit in not looking at sharescur(0). After all, nothing else you're using when looking at a company's fundamentals reflects post-filing changes. So if a company buys back some shares four weeks after its filing and you use sharescur(0), your P/E will change but your ROA and profit margin won't.


I got burnt big time a few times. There were stocks showing an owner earnings yield of 33% when their true yield was less than 5%. The rest of the market knew.

You may not run into this issue too often because you use multiple factors for each stock. But subscribers like me use some systems that are pure value and we need a way to fix this issue.

nothing else you're using when looking at a company's fundamentals reflects post-filing changes.

Changes in margins don't give everyone else an edge over us because we get the data about the same time as everyone else. In fact, we are usually able to act on it before human analysts, so that gives us an edge.

However, changes in share counts can take four and a half months until they get reported in the financial statements. The rest of the market knows right away. Value factors can be thrown off by an order of magnitude. That's currently a huge weakness in the data we have been using.

This issue may partially explain the struggles many have had with value ratios since the switch from Reuters in 2012 (when this was not an issue). I am hopeful that FactSet has the data to fix this issue.

May 26, 2020 11:00:12 AM       
Edit 1 times, last edit by Chipper6 at May 26, 2020 11:02:20 AM
yuvaltaylor
Re: New engine released

There is some merit in not looking at sharescur(0). After all, nothing else you're using when looking at a company's fundamentals reflects post-filing changes. So if a company buys back some shares four weeks after its filing and you use sharescur(0), your P/E will change but your ROA and profit margin won't.


I got burnt big time a few times. There were stocks showing an owner earnings yield of 33% when their true yield was less than 5%. The rest of the market knew.

You may not run into this issue too often because you use multiple factors for each stock. But subscribers like me use some systems that are pure value and we need a way to fix this issue.

nothing else you're using when looking at a company's fundamentals reflects post-filing changes.

Changes in margins don't give everyone else an edge over us because we get the data about the same time as everyone else. In fact, we are usually able to act on it before human analysts, so that gives us an edge.

However, changes in share counts can take four and a half months until they get reported in the financial statements. The rest of the market knows right away. Value factors can be thrown off by an order of magnitude. That's currently a huge weakness in the data we have been using.

This issue may partially explain the struggles many have had with value ratios since the switch from Reuters in 2012 (when this was not an issue). I am hopeful that FactSet has the data to fix this issue.


It's not a question of not having the data. We have the data; FactSet and Compustat have the data. It's the discrepancy between the number of shares on the market and the number of shares reported in the most recent financial statement. SharesCur(0) is the number of shares on the market; Shares and SharesFD are the number of shares reported. There's no one right way to combine the two: they're very different numbers. I think that using my formula or something like it will get you what you need.

Yuval Taylor
Product Manager, Portfolio123
invest(igations)
Any opinions or recommendations in this message are not opinions or recommendations of Portfolio123 Securities LLC.

May 26, 2020 5:33:32 PM       
Chipper6
Re: New engine released

It's not a question of not having the data. We have the data; FactSet and Compustat have the data.

I didn't quite get that. If you have the data, why don't you expose it?

It's the discrepancy between the number of shares on the market and the number of shares reported in the most recent financial statement. SharesCur(0) is the number of shares on the market; Shares and SharesFD are the number of shares reported. There's no one right way to combine the two: they're very different numbers. I think that using my formula or something like it will get you what you need.

I'm not sure what you mean here. As you may know, SharesCur(0) is up to date but is limited to a single share class. In most cases they should probably be exactly the same.

I find it hard to believe that you don't have the data for up to date share counts across all share classes. Reuters had it. Many financial sites have it. Doesn't Compustat and Factset have it too? In my mind this is basic market information. I don't like to trade against people who have better information than me.

Your formula is an estimate. I have used a similar formula myself, but I have found material errors in it to the point that it is not any more accurate than Shares from the financial statement. Perhaps the problem is that it assumes that the ratio of shares in the various shares classes remained constant. This may not be true when new shares are issued. I have abandoned my formula because of those errors.

I am looking for the actual up to date share counts.

May 26, 2020 11:25:02 PM       
yuvaltaylor
Re: New engine released

It's not a question of not having the data. We have the data; FactSet and Compustat have the data.

I didn't quite get that. If you have the data, why don't you expose it?

It's the discrepancy between the number of shares on the market and the number of shares reported in the most recent financial statement. SharesCur(0) is the number of shares on the market; Shares and SharesFD are the number of shares reported. There's no one right way to combine the two: they're very different numbers. I think that using my formula or something like it will get you what you need.

I'm not sure what you mean here. As you may know, SharesCur(0) is up to date but is limited to a single share class. In most cases they should probably be exactly the same.

I find it hard to believe that you don't have the data for up to date share counts across all share classes. Reuters had it. Many financial sites have it. Doesn't Compustat and Factset have it too? In my mind this is basic market information. I don't like to trade against people who have better information than me.

Your formula is an estimate. I have used a similar formula myself, but I have found material errors in it to the point that it is not any more accurate than Shares from the financial statement. Perhaps the problem is that it assumes that the ratio of shares in the various shares classes remained constant. This may not be true when new shares are issued. I have abandoned my formula because of those errors.

I am looking for the actual up to date share counts.


We do expose the data. The SharesCur(0) data comes from ICE. It's up-to-date and changes constantly. The financial statement data comes from Compustat and FactSet. It's all there. It's up to users to put those together in the manner that suits them best. I'd like to see an example of "up to date share counts across all share classes."

Take, for example, ZM. SharesCur(0) is 167.56. SharesCur(40) is 127.47. SharesQ is 278.73. SharesFDQ is 292.79. On which of those numbers do you base Zoom's market cap? At $164 per share, Portfolio123 has the market cap at $45.71 billion, which is what you'd get if you used SharesQ. Google Finance, Fidelity, and Thomson Reuters all base their market cap on SharesFDQ, which is a tiny bit higher. But Zoom has well over 40 million more shares in circulation than it did when it released its last financial report on March 4. Should its market cap really be $59.79 billion? I don't know. Do you? Is there another website that says so? Is there a website that currently gives it a share count of over 300 million, which may be what it's going to report next week? I'm not trying to be contentious. I'm honestly very curious about this and have no idea what the right answer is.

Yuval Taylor
Product Manager, Portfolio123
invest(igations)
Any opinions or recommendations in this message are not opinions or recommendations of Portfolio123 Securities LLC.

May 26, 2020 11:51:19 PM       
Edit 4 times, last edit by yuvaltaylor at May 27, 2020 8:52:27 AM
Chipper6
Re: New engine released

Take, for example, ZM. SharesCur(0) is 167.56. SharesCur(40) is 127.47. SharesQ is 278.73. SharesFDQ is 292.79. On which of those numbers do you base Zoom's market cap? At $164 per share, Portfolio123 has the market cap at $45.71 billion, which is what you'd get if you used SharesQ. Google Finance, Fidelity, and Thomson Reuters all base their market cap on SharesFDQ, which is a tiny bit higher. But Zoom has well over 40 million more shares in circulation than it did when it released its last financial report on March 4. Should its market cap really be $59.79 billion? I don't know. Do you? Is there another website that says so? Is there a website that currently gives it a share count of over 300 million, which may be what it's going to report next week? I'm not trying to be contentious. I'm honestly very curious about this and have no idea what the right answer is.

Zoom is an interesting case and illustrates why we need an up to date share count. The company has been busy printing shares at this high price.

It currently has at least 319.09m basic shares outstanding, which makes its market cap at least $51.68b as of yesterday's close of $161.97. This is not counting options or any potential increase in class B shares since March 6. SharesQ, or even SharesFDQ undercounts the number of shares outstanding by a lot. This would make value ratios look much better than they are.

Sources? The 10k, which notes [EDIT: on the first page or so] that basic shares outstanding was 279m as of Mar 6, and was comprised of 127.47m class A and 151.53 class B shares, and SharesCur, which tracks class A shares and has gone up to 167.56m.

Yahoo Finance seems to track the Mar 6 figures. [EDIT: Why do I say that? Because] it is showing a market cap of 45.668B. At a current price of 161.97, that implies a share count of about 280m, which was the Mar 6 figure. This is more accurate than SharesQ or SharesFDQ, but not as accurate as the numbers above.

Basically, Zoom is a good example illustrating why we need up to date share counts.

May 28, 2020 12:44:51 AM       
Edit 4 times, last edit by Chipper6 at May 28, 2020 9:24:03 AM
yuvaltaylor
Re: New engine released


Yahoo Finance seems to track the Mar 6 figures. [EDIT: Why do I say that? Because] it is showing a market cap of 45.668B. At a current price of 161.97, that implies a share count of about 280m, which was the Mar 6 figure. This is more accurate than SharesQ or SharesFDQ, but not as accurate as the numbers above.

Basically, Zoom is a good example illustrating why we need up to date share counts.


Yahoo Finance shows a market cap that's $2 billion higher than ours because they're using SharesFDQ and we're using SharesQ. Nobody is showing market caps that incorporate the increase in SharesCur since the 10-K filing. We could do so but we'd be a complete outlier and our market caps would differ radically from everyone else's. There are industry standards that we need to adhere to.

The other question is this one: how do you know that the increase in SharesCur reflects the issuance of new shares? See the screenshot below. Last December, Zoom's share count increased by about 4 million while their SharesCur increased by about 55 million. A similar thing happened in September. The only way to explain that discrepancy is that Zoom is issuing a lot fewer new shares than the number of closely held, treasury, or newly converted shares that are being brought onto the market. And those would not affect the market cap.

Attachment Screenshot_2020-05-28 Fundamental Chart - Portfolio123.png (87756 bytes) (Download count: 60)


Yuval Taylor
Product Manager, Portfolio123
invest(igations)
Any opinions or recommendations in this message are not opinions or recommendations of Portfolio123 Securities LLC.

May 28, 2020 12:23:00 PM       
yuvaltaylor
Re: New engine released

FYI, we've now changed the FactSet formulas for Cost of Goods and Gross Profit. You'll see that for banks and insurance companies, these items are N/A. There is simply no good way to get these numbers for banks and insurance companies, which don't report these items, and what Compustat was doing to get them wasn't kosher, in our opinion, and not possible to replicate anyway.

Yuval Taylor
Product Manager, Portfolio123
invest(igations)
Any opinions or recommendations in this message are not opinions or recommendations of Portfolio123 Securities LLC.

May 29, 2020 11:04:08 AM       
Edit 3 times, last edit by yuvaltaylor at May 29, 2020 10:46:13 PM
Posts: 90    Pages: 9    Prev 1 2 3 4 5 6 7 8 9
 Last Post