Show HN: I built a website to create financial models for any stock online
useequityval.com
useequityval.com
Besides, the comment wasn't unkind to the present poster and is actually kind to future posters, so the admonishment and down-votes seem unwarranted.
I'd want to be able to define a model like "I think the fair value of a company is its net current assets plus its last 5 years of earnings", and have the tool email me whenever I can buy shares in any company substantially below that price, or sell them substantially above.
It wouldn't be very hard to do, it just needs the data.
Where are you getting your data from?
Be careful with investing using a model like this: you're basing the value of the company entirely on its historical performance and not at all looking at what its future may look like. It's a great way to end up in value traps.
Exaggerated example, you see Intel's stock price has taken a huge dive that puts it under your limit so you buy a bunch of the stock. But it turns out the reason Intel is now cheap is because ARM, AMD and NVidia have eaten its lunch, and its future looks incredibly shaky.
Likewise, selling stocks that are substantially above some past valuation is also a good way to exit from positions that are likely to continue to outperform. Stocks get "re-rated" by the market when their future prospects have improved. That's not a great time to exit the stock - if anything it's a time to buy more.
YMMV, value investing is still successful of course, but you do need to look forwards not backwards.
I don't see attribution on useequityval.com?
Edit: Chipotle must be headed for E. Coli again: Projected Price: $-2,890.64
Where did you find this? None of the data in https://fred.stlouisfed.org/categories/32440 confirms this other than the small bank credit card delinquencies.
For Apple it is whether they can stay a prominent platform that can capture a significant portion of the value created on it.
For Tesla it is how fast their self driving tech will evolve.
For Google it is whether AI will make search obsolete or more valuable.
Everything that Buffet says falls into one of two categories: [1] He has no relevant knowledge. [2] He/Berkshire makes money from it.
[2] Is the most common. (His railways transport oil. His insurance companies make money from the inheritance tax, a tax that his estate won't pay. Etc.)
> His insurance companies make money from the inheritance tax
There are lots of investments that are way more boring than Apple, Tesla, and Google, and that should be analyzed with DCF.
DCF is more for slow and steadily growing companies that might be undervalued for just a moment because of factors outside of the company's control.
Tech companies can be harder to value via DCF, because as you say, there are monumental factors that outweigh historical cash flow trends.
This is part of why Warren Buffet historically avoided tech. It is wildly unpredictable from a pure historical cash flow numbers standpoint.
DCF is not a crystal ball.
The types of questions you're talking about aren't well analyzed by DCF.
But something like the fair value of Walmart can absolutely be determined via DCF.
You expect that investing in all of the companies in the upper half of the list will result in a better long-term return than investing in the lower half of the list?
This is THE way fundamental analysis is done. There are many ways to model - but this is the most robust as you have the most inputs.
Maybe people just have no idea how “fundamental analysis” works or even what it means?
This is how it works. That doesn’t mean you’re going to beat the market, or even make a positive return. But it’s essentially the basic fundamental analysis model.
- price (absolute level)
- price movements (relative change)
Your claim is that an equity's price isn't usually a reflection of fundamental value.
On how many days this year (or this decade) was Google's market cap lower than that of Cheesecake Factory? Why?
AIUI Rentec holds positions for minutes or days (or maybe even weeks?), not seconds or less.
My general point still stands.
"Hey, the model starts with the previous years growth numbers projected out 5 years. You are meant to give your assumptions to get to a valuation. NVDA revenue grew 125% last year so projecting that out for 5 straight years results in a very high value but is most likely inaccurate!"
I should probably take nvda off the home page lol
What stack did you use? Im trying to learn more about web development and was in the process of learning how to make my idea. Im leaning towards the MERN stack rn but not exactly sure if its a good idea, so im just asking you also what were the criterias/questions you asked yourself for choosing your tech stack?
Besides, anything aviation is a financial accounting shenanigans party. Many airlines, for example, are "loss leaders" - the money actually comes from their "reward programs" and credit cards [1], a practice that's been going on for decades now [2].
[1] https://www.theatlantic.com/ideas/archive/2023/09/airlines-b...
[2] https://airlinegeeks.com/2021/12/17/here-s-why-airline-loyal...
Try something like this in your global.css https://pasteboard.co/VAf6hbqgxOoU.png
It's simply a way. There are "plenty of ways to skin a cat". Best is highly debatable. I'd argue best is relative given the dev's experience and interest. Might be best if the dev was you.
Also, third Show HN submission so persistence and faith paid off..
The models that make the big bucks are the ones that ingest a ton of other external data to predict the numbers that go in the data entry cells here. And yeah, those don’t get posted online for free.
If you look at the Price Targets for many "growth stocks", you'll see that professional analysts struggle with this as much as anyone.
Even for stocks with predictable profit models like resource extraction companies, it can be near useless because as soon as some external catalyst people didn't expect comes along triggers the commodity price to change significantly, the DCF valuation goes out the window.
IMO DCF is more useful for sensitivity analysis than actually valuing a business in some "accurate way".
All that being said -- it is definitely valuable to understand where price targets come from!
Also, I think it'd be great to set your defaults assumptions for a stock to breakeven pricing, and allow users to adjust from there.
I see a tool like this being useful, but I think positioning it as a deep valuation tool is much less useful in the mindsphere of investors as an elite pen and napkin, which I think people would use.
Do such models even exist? As far as I know NN are only good at function approximation. Why do we think that the market evolves according to a mathematical function? To me, financial markets seem chaotic in nature.
But the noble thing to do is to stop luring people into these wealth-gap increasing casinos.
... But with one year lag.
Current Price: $56.22
Projected Price: $20,806,772,549,824,028.00
Difference: 37,009,556,296,378,460%
Glad I'm long.
Assuming that a revenue growth rate of 84,762.39% is (a) a valid number and (b) expected to remain the same over the next X years does not quote-unquote "make sense"
That is a decent return. I bought a call option for RILY. Went to zero. Well...
Edit: why disagree?
Also there is no real company / person mentioned as an owner at least on the website so the privacy policy is kind of a "trust me bro".
If it was even remotely reliable, this would be a billion dollar idea (at least) treated as a secret sauce in an investment company.
Instead, it takes some input, throws some garbage output, but as it's not blatantly just random numbers, people think it's helpful for investing.
You could probably get away with charging a subscription fee.
So, your tool sounds interesting, but I don't know enough to use it. Do you have a tutorial video or anything that might walk me through a user story on how somebody would use your tool? Or, do you know of a good fundamentals course that would give me the knowledge to be a customer of your tool?
A few "trading api" shops did this last time I looked into it.
You're assuming that every human on earth is extremely selfish.
If what you say is true, there would be no teachers in the world. Because most teachers make less than the students they teach.
There are unselfish people out there.
NVDA revenue grew 125% last year so projecting that out for 5 straight years results in a very high value but is most likely inaccurate!
OP: What are “AI Models"
So you can say, "I think AAPL will have strong growth over the next 5 years, they will also see costs go down as xyz will happen" and the AI will use your assumptions to fill in the DCF