Sorry everyone! And thank you! Due to high volume I am hitting my API limits. If you are getting a server error that is why. The limit resets ever minute so please wait and try again.
Consider caching your API results, and eventually upgrading to a paid tier to increase your API limits. Also, I should be able to enter a stock ticker on your website and see the corresponding model without needing to sign up.
My tip is to re-evaluate why you think you are getting downvoted to better understand your tone and wording rather than assume the downvotes are not warranted.
Very nice, coincidentally I have been thinking about making a similar tool lately, but I'd make it quite a bit more conservative.
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.
> 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.
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.
> Attribution is required for all users. It is as simple as putting “Data provided by Financial Modeling Prep” somewhere on your site or app and linking that text to https://financialmodelingprep.com/developer/docs/.
Getting the data is always the hard part for stock apps. My experience has been that it’s very difficult to get up to date data for free. You’ll probably have to pay for an API. You’ll probably want to anyway so that you aren’t juggling several free APIs to get the same data, or so you’re working with a mature, easy to use, RESTful API. I’ve worked with stock market APIs that are miserable to use.
I don't know if this is the reason but it is somewhat trivial to roll your own. This is actually close to the story of how I became a programmer in the first place. 17 years ago or so, I made something like this in Excel and figured out how to populate the data from the Yahoo! Finance API, then learned about FRED, BLS, other sources of possibly relevant economic data, but still doing everything in Excel, eventually deciding to learn C++ since I'd heard it was the industry standard for finance. I caught the bug, ended up being more into software than finance specifically, and the rest is history. My motivation at the time was this was the middle of the big global financial crisis, everything had tanked, and it seemed like a great time to get in low. Buying my first house in 2009 was probably the most advantageous thing I did, though I was at least able to do that by tax-free early liquidation of an IRA that was full of self-picked stocks.
It's predicting a contraction in value of 20% to 70% of every stock I throw at it. 40k DIJA is a more-than-inflationary bubble popping within a few years?
Edit: Chipotle must be headed for E. Coli again: Projected Price: $-2,890.64
I mean all USA debt delinquencies are currently between the same levels as Q1 and Q2 of 2008... foreclosures aren't near as high, but they are trending up quickly.
Im skeptical whether these types of models can help you make better investment decisions. The fate of a company usually hinges on some fundamental question.
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.
Yes - it is well known there is no "secret investing sauce" - this has been proven over and over. The only people who tell you otherwise are people with something to sell you, or people who have tricked themselves into believing otherwise. This kind of stuff pops up whenever the market is on a tear.
The secret sauce is simple: low cost ETFs that track the S&P 500 index. Regular, automatic investments. Go to sleep. Wake up 30/40/50 years later. You will be rich. Warren Buffet supports this idea for retail investors.
> Warren Buffet supports this idea for retail investors.
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
I don't follow this. Can you explain more? Buffet has been quite clear that he expects low cost index-tracking ETFs to outperform even hedge funds with 2&20 fee structure.
Let's say you sit down and make a DCF for every company in the S&P 500. Now sort them by how much their current price differs from what you call "fair value".
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?
I’m not sure why you’re being downvoted, because it’s true.
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.
Yeah its an art not a science for sure. Those questions and how you think the answers will play out should inform the inputs you use for your model. DCF modeling is not a silver bullet to predict a stock price thats for sure.
I'm not sure why anyone would invoke Renaissance as an example of anything, giving how little people know about what they concretely do. Seems ironically unscientific of people who admire Jim Simons.
I was referring to an equities price generally being in a wide band channel of price over a year. I was not suggesting an equity goes wildly below it's intrinsic value often (although it is not uncommon still)
…yes. Which is why OP made the argument that this tool, which helps with fundamental performance, isn’t very useful for calculating an equity’s price target.
"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 think maybe come up with one or two more simple, but dumb models. Show all three at once to demonstrate the wildly different results that can come out given different assumptions.
Is it still in Beta? Seem to have quite a few bugs, not letting me pick a stock from the list unless I refresh the page. Besides, it would be better to have some info about how was the projected price calculated.
I would love to see some tooltips or something to explain the relevance of some of the stock metrics and what a good vs bad number would look like. Something as simple as "Higher number = better" would be useful.
Hi, loved the idea behind your product. Just a quick question about your tech stack?
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?
I wish it had the ability to statistically correlate operating expenses or capital expenditures, for instance, with share-prices, especially for historical "backtesting", to determine if there is a positive or negative correlation.
That's where this company should end up in any fair world, given all the evidence pointing to intentional misconduct and structural governance issues as a cause for multiple fatal crashes and other serious incidents. Snark aside, this may just be a case of Boeing's numbers being distorted by its business model and airlines paying only on delivery of airplanes (which has been seriously cut back due to regulators imposing production stops and caps as a result of all its issues for years now), but Boeing having to pay their vendors upon receipt of parts.
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].
Hi, DCF models are not very effective when a company is cashflow negative. That can result in weird looking values. On the last tab of the model you can switch the exit multiple to look at EV-Sales which will work better on a cashflow negative business like Boeing. Doing this with the model defaults results in a value of $149. Thanks for checking it out!
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.
I’m consistently surprised at the lack of basic financial literacy in the HN crowd. The title says this is a tool to create financial models, not that it is a well-tuned financial model. Yeah, you have to make better assumptions than straight-line extrapolation from last year’s trend to arrive at reasonable valuations. The point is that this shows you what assumptions you have to make to arrive at a rationally-calculated stock value, and pre-populates all of the other necessary data and formulas in a slick UI, which is pretty cool.
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.
Mmm. The one thing I mostly learned from doing DCF models myself in Excel was how sensitive the output was to the different inputs. It didn't really help me to do the modelling itself, the hard part is predicting the future, not the Excel formulas.
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!
Any way do regional and or segment breakouts? The way I see your tool, it's a great tool for the back of the napkin-ers but as it is right now, it doesn't obviate modeling because you're still going to have to do the breakouts in excel.
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.
You'd be surprised to learn then that most PE deals run on the back of literal pen and napkin models. In our process at my former firm, one of the largest megafunds out there, we would scout and model new opportunities using elaborate models, report them to the MDs who would then literally whip out a notepad or a napkin mid-presentation to decide whether the investment had potential. If it was in the grey zone, it was on us to demonstrate that it had merit, using more complicated modelling and analysis.
That doesn't surprise me at all, I've heard it's the same in private credit. I think a lot of the initial ground work that's done in due diligence and modelling is more to flush out any red flags than to inform the investment decision itself, and a model is only as good as its assumptions anyway...
Spot on. We used a bunch of screening processes to mine opportunities, then use models to weed out the duds, or to buttress our investment thesis before presentation to the MDs.
>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.
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.
It’s like asking why are gym memberships and diet advice more marketed than eating less and free body weight exercises you can do anywhere. There is a market for solutions and good news, whereas there isn’t one for bad news, even if it’s free.
Yes but it's not really about the number crunching itself (NN's or something else), but more about modelling the "wider world" ("macro") outside the company that could drive its business. Like if you wanted to project an oil company's cash flows into the future, you'd do a lot of analysis and number crunching on things like IEA projections, known oil reserves, risk scenario analysis and the like.
It highlights an education opportunity for OP. Some people think this is a proprietary financial modeling system because they don’t know what a DCF is.
I feel like OP would have been better off not pre-populating the models, just to make the point of the tool clearer. A lot of folks here seemed to have missed the point of the tool.
The hard thing about modeling is not the math to get to present value of the stock. It's figuring out which assumptions make sense.
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"
Haha the initial model is not a good representation of value as it projects the lasts years metrics out 5 years. 80K% growth for 5 years would do that. Thanks for pointing it out. I need to incorporate some boundaries for the initial values lol.
I put in quite a few stocks and the results were often strange. Negative values, under a dollar, etc. These were all stable companies, not penny stocks.
To be fair to the creator, he lets you put in your own assumptions via the revenue growth fields (which is what's skewing your projections). That's frankly better UX than one of its competitors, which does not let you do that.
The concern isn't that they would invest bed on obviously wrong numbers. The concern is that someone invests based on believable numbers that are incorrect. There's no information on accuracy measures or verification. Offering this model to others could be an SEC/FINRA violation.
Likely the people who downvoted have no experience with the legal part of stock related discussions/tools (implied expectations/suggestions, etc), there is a reason why you see "not a financial advice" and other legal texts presented in videos discussing stocks.
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".
Two economists were walking down the street. The first economist spotted a £20 note on the ground and stopped to pick it up. The second economist asked "why are you stopping?". The first economist said "to pick up this £20 note". The second economist replied "don't be stupid, if there was a £20 note on the ground somebody would already have taken it". The first economist considered this for a moment, then nodded, and they both walked on.
I'm interesting in learning more about individual stock investing. I mean, I've been doing it for years, but more just on hype. I've been very successful, but I do very little research on fundamentals and I want to do more of that.
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?
It sounds like a couple people in this thread have misunderstood how to use your tool. Perhaps you want to hide the projected numbers at least until they start to put in their own assumptions.
ah that makes sense. thank you for clarifying. i was expecting a fair valuation based on YOUR assumptions and the value I misconstrued was that YOUR assumptions are better than mine so I must pay you.
lol I hope so to! there is no AI on the model though. Pro users can have an AI fill in the model for them. But the base model you see when you load up a stock is just projecting the previous years metrics out 5 years!
Everyone: This is a nice simple interface for making a discounted cash flow (DCF) model, and the pro version lets you export that to Excel. There’s still a lot of work to fill in those fields. More on DCF:
https://corporatefinanceinstitute.com/resources/valuation/dc...
Hi, the pro version allows you to tell an AI what you think aboout the company and it will fill in the DCF model for you.
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
When something isn't good, you needn't pretend that it is, but don't be gratuitously negative.
But more importantly, what makes your comment shallow is not the ostensible depth and heft of your links, it's that it shows little evidence of having looked at the thing being showhn - it's a thread-invariant trope.
If someone does a Show HN for a Twitter for pets you are not required to applaud it, instinctively or otherwise, but you are asked to not reflexively dismiss it in thread comments. Critique is fair game, of course.
> Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that".
I didn't comment on whether you've read the article. There's a strong expectation a toplevel comment in a Show HN thread is concretely about the thing being showhn. That's not some complicated thing that requires the subtle parsing of multiple rules.
Yes, and you quoted yourself that just because it’s a “ShowHN” doesn’t mean that it has to be praised effusively and get a attaboy.
My negativity wasn’t “gratuitous”. His tool was meant to predict stock price. Every reputable article shows that kind of analysis is foolhardy- or as the WSJ put it - about as good as blind monkeys
I think the key thing is that it was a straightforwardly poor Show HN comment. What Specific Word of The Guidelines It Hath Not Transgressed Against as well as Who Is The Real Guidelines Sinner are both not particularly important.
Responding with personal attacks is not a good look, especially when you're using slurs (as 'special' is, when used in this context).
It doesn't matter how bad or inapplicable the criticism is. If you do this, they've brought you down to their level, and regardless of who wins the resulting argument, you've lost.
