This analysis is focused on my continuing journey to gradually off load my stock analysis processing power onto neural nets where this net attempts to forecast the value of AI stocks 3 years out based on my EC indicator resulting in a upwards pressure under the stock price indicator.
- AI Stocks Value Forecaster (ASVF).
- How I Use ASVF6 - Percent Upwards Pressure (PUP)
- AI Stocks Buying Levels Plus ASVF & PUP
- AI Stocks Portfolio Buying Levels
- Dow Stock Market Trend Forecasting Neural Nets
- Pattern Recognition
- Trend Analysis Preprocessing
- Crossing the Rubicon With These Three High Risk Tech Stocks
- Cheap Chinese Tech Stock 1
- Cheap Chinese Tech Stock 2
- Cheap Chinese Tech Stock 3
- CME Black Swan
AI Stocks Value Forecaster (ASVF).
My initial attempts some years ago at feeding raw financial data such as PE ratios, EBITA, PEG into neural nets failed to deliver anything useful, clearly it was not going to be that easy! What's required is to preprocess the data to make the various inputs better gel together rather than producing noise which means input from human intelligence (me) to preprocess the data before training the neural nets, and so was born what I call the Expensive / Cheap (EC) Indicator that includes 15 stock financial indicators all fine tuned manually to resolve in better being able to discriminate between stocks in fundamental valuation terms, add price to that to become 16 inputs. that have been preprocessed as per their formulation into the EC indicator.
IMPORTANT POINTS REGARDING USE OF THE ASVF
1. That it is trained on bull market price and EC financial data, so it is unknown how it would behave during a bear market, probably over estimate future outlook for a stock prices, but it's unknown as there is a lag between price movements and corporate financial data. I have thrown bad numbers in and it does still tend to project higher prices but to a far less extent, but it does have a bull market bias.
2. That it is unknown how accurate the valuation forecasts will turn out to be as it is not actually forecasting price i.e. trend analysis, but price in terms of valuations. How much a stock will be worth in relative terms compared to what has come to pass under it's EC values..
3. That the data is limited to the stocks that I generate an EC for until I get around to writing better routines to auto generates the data to calc the EC with error checking..
4. That the readings are volatile i.e. there is very little smoothing of the forecasts i.e. if Google fell today by $400 then the forecast would fall by about roughly $400 to track the price lower, if it rose by $400 then the forecast would go up by about $400, So it is definitely not a trading tool but rather a relative future prospects valuation tool.
5. That it is still developmental i.e. the EC indicator is constantly being refined which means that preprocessing for inputs changes which can effect the forecasts i.e. I may discover a flaw in the formulates that requires a rethink, though which was the whole point of creating the EC indicator as if the EC indicator is useful than the preprocessing should also prove useful.
6. The key here is preprocessing the data, that it is always being fine tuned as per the construct of the EC indicator.
How I Use ASVF - Percent Upwards Pressure (PUP)
(Note all data in the AI / buying levels calc's are based on data upto 14th July 2021)
In my opinion the most important use for this tool is in the 2nd new column on my table AI stocks portfolio table, the Percent Upwards Pressure (PUP) which is basically the ASVF divided by the current stock price for percent change.
What this shows is literally the amount of UPWARDS pressure a stock is currently experiencing as the price as it is being PULLED towards it's future forecast which to me is far more useful than an always distant 3 year price forecast.
For instance it tells me which stocks are likely to have mild or no maybe even no price corrections due to the fact they have a high degree of upwards buying price pressure which in market terms translates into a LOT of investors waiting on the sidelines for the price to drop so that they can BUY! So what happens? The price does NOT drop instead it continues trending higher frustrating waiting to buy investors.
This is one of the key problems we have when trying to buy a deviation from the high. What do we do if the stock FAILS to fall? FAILS to see sense and follow the other stocks lower. Well we now get to peak under the markets hood and have advance warning of stocks that are going to be strong in relative terms come what may.
So for instance here I am sat waiting for Nvidia to correct towards my buying levels after selling all of my holdings at $715, having ignored the AI's advice that it would be a bad idea to SELL Nvidia stock, and what did Nvidia do it continued rocketing higher to well above $800 and is parked at $820 as I write (14th July 2021), and even after updating the EC levels the AI is still telling me (+171%) that it is not a good idea to sell Nvidia.
But what is of use right now is the Upwards Pressure Indicator.
It gives an alternative opinion on whether I should be buy, sell or hold an AI stock, an independant indicator. However, definitely not an automated trading system!
What I am actually looking for is if the indicator is telling me something significantly different, so that I can take a deeper look at that stock.
Whilst Yes having a distant price target can be comforting, but 3 years is too LONG! Especially for a primitive AI! And especially as we all know that company reports can be suspect, that can show huge swings between quarters especially at market junctures, so I would not take the 3 year valuation forecasts too seriously but instead focus on PUP.
AI Stocks Buying Levels Plus ASVF & PUP
Google $2620, ASVF $4396, PUP 168% WHITE CHAPEL AI Dominance!
Google is the primary stock on my list, the Neural Net (NN) confirms why it will remain so many years before I see a reason to completely disinvest from this stock, despite being up 50% year to date the stock has a high upwards pressure under it of 68%.
Google looks set to spread it's AI tentacles into the hardware space which means in the not to distant future it's going to be competing against the semiconductor tech giants for market share. Google AI's initial chip market to disrupt will be smartphone's in many ways similar to what Tesla is doing with autonomous vehicles right now Google will soon be seen be doing to smartphone's as Google's in house White Chapel chip that in hardware spec terms is LESS sophisticated than Qualcom's Snap dragons that it seeks to compete against, which on face value implies to expect middle of the road performance given its 8 core specs. However the key difference is that the chip has been designed by AI and is in a constant state of revision, Google AI continuously improving the design where today's traditional processor generations that extend from 12 to 18 months cut down to only mere days but even hours! This gives you a taste of what is coming and why the likes of Qualcom and Apple amongst others should be worried though also have an increasing input of AI in the design of their processors just not to the extent that Google has.
What this means is I will take advantage of short-term downwards stock price volatility to load up with Google stock. Of course when investing one needs to keep an eye on valuations for that is what ensures the probability for successful investments i.e. buying at low valuations and selling at high, but Google looks set to continue to grow its earnings well into the distant future. It;s just a question of which corporations that are slow to react will pay the price.
So as things Stand Google remains my primary AI stock to be invested in as it continues to demonstrate its lead as it trends towards AI dominance. Even so right now I cannot pay 35 times earnings so will either wait for earnings to play catchup and look to buy Google stock at under 28X earnings.
AMAZON $3677, ASVF $4320, PUP 17%
With only 17% upwards pressure confirms that I was right to de-risk by selling ALL of my Amazon stock. Similarly poor upwards pressure translates into under relative performance ( Up 24% YTD). When the market corrects, I now even more so than before expect Amazon to correct hard. Neither I nor the AI sees much justification for the stocks current surge higher, and my long standing buying level of $3020 may end proving to be significantly higher than where Amazon could trade down to.
Microsoft $281, ASVF $517, PUP 84%
Microsoft gets a stronger reading than my perception of the stock i.e. as per my last analysis I saw it as a middle ranking of the top AI stocks, hence sold 50% of my holdings, instead the AI see's the stock as having far greater upwards potential and support under it, even more than Google!
Apple $145.6, ASVF $253, PUP 74%
I have not been a fan of Apple's prospects relative prospects since at least November 2020 i.e. expected under performance which has transpired as the stock is only up (+10% YTD. But the AI is giving Apple benefit of doubt saying that it may not have much downside during a correction, so I don't agree with the AI.
Facebook $352, ASVF $749, PUP 113%
Facebook along with Google for a while have been my two best AI stocks., Unfortunately it does look like I made a mistake of selling 70% of my holdings, since which the stock has risen as indicated it would by the AI due to continuing immense upwards pressure of 113%. Though I don't see myself accumulating any Facebook stock above $298. Which could come back to bite me if Facebook fails to correct to below $300, if so next time I will take better note of the ASVF.
Nvidia $820, ASVF $1433, PUP 75%
The AI is effectively saying I was wrong to sell ALL of my holdings at $715 as it prices future prospects for the stock on par with that of Google, which I don't agree with. Anyway I remain uber bearish on Nvidia that I see as floating on thin air, even to the extent of going SHORT on the stock (with a tight stop)! So will ignore the ASVF in this case.
AMD $90.3, ASVF $192, PUP 113%
AMD has gone nowhere year to date but the AI is saying it has huge upwards pressure which means corrections should be milder than for others. Which concurs with my expectations Though it looks like I can forget about my fantasy buying level around $60, instead it looks like I will be a lucky to get a dip to below $80! Definitely not a stock to exit, which luckily I did not sell any of during my recent selling binge.
Samsung $1750, ASVF $2340, PUP 34%
Samsung after the stock price surge off the March 2020 low to the recent 2070 high appears to be going back to sleep, 33% upwards pressure is not much, the AI suggests its probably best to move Samsung to the sleeping giant category. Still it is yielding 3.,77% whist we wait for this sleeping giant to wake up again!
TSMC $124, ASVF $220, PUP 78%
The AI rates TSMC as a middle of the road AI stock, which fits my overall view of the this semiconductor stock that basically has very little effective competition for it's fab's right now, but fab plants ARE being built hence future prospects are more measured..
JNJ $169.3, ASVF $277, 63%
The AI confirms that JNJ should continue trending higher out of its preceding trading range.
IBM $90.3, ASVF $177, PUP 26%
The AI agrees with my decision to sell ALL of my IBM stock towards the top of its trading range. The AI does not see much upside potential for IBM. I am little more hopeful though whilst it is in it's trading range I will continue to sell at the top with a view to buying towards the bottom. Though I remain more optimistic for a whole host of quantum reasons on IBM in the long run.
INTEL $56.9, ASVF $151, PUP 166%
INTEL the literal sleeping giant that I have repeatedly flagged as being one of the few good buys out there right now, and the AI confirms this by giving it the highest upwards buying pressure of all the stocks so far at 166%. And as I have stated several times that I would not be surprised if Intel in a few short years time is trading towards $200. which the AI is confirming, I will definitely buy more Intel on a dip below $50.
Roche $47.5, ASVF $72, PUP 52%
A PUP of 52% is not that bad for this Big Pharma as the sector as a whole is unloved, maybe a sign to accumulate into pharma stocks during a correction.
GSK $40.5, ASVF $54, PUP 34%
Why am I still holding GSK? Oh yeah, the 5% dividend yield, something to remember when just looking at the stock charts. Dividends have delivered an effective 80% gain (against buying price and if reinvested) over the past 10 years so not quite as bleak an investment as it first appears. Still the AI confirms not to expect much upside for the stock price and so I may put a limit order in to sell towards the top of its' trading range.
The bottom line is that Amazon is a ticking time bomb for it's stock investors whilst Intel is the hidden gem.
This also makes me wonder if most of us are misunderstanding Tesla and why the PE ratio for Tesla could be mostly irrelevant. Yes Tesla is an automaker but what I am increasingly becoming aware of is that Tesla has an exponentially increasing driving dataset, each of it's Tesla models has a myriad of sensors that transmit data back to Tesla, so forget the dozens or even hundreds of cars that others in the self drive game have deployed, Tesla has over 1 million cars passively generating DATA for Tesla! No other corporation comes anywhere close to that and so that is the REAL value of Tesla! Though still I would not consider buying north of $300.
AI Stocks Portfolio Buying Levels
Overall what the AI is stating is not impacting on my buying levels much as I still expect stocks to correct lower imminently and thus I will be looking to buy near the buying levels mentioned, where at the back of my mind is not that stocks won't correct, rather that I may end up buying too high! As per the Financial Crisis focus of my previous in-depth analysis.
We are in a dangerous market where FOMO (Fear Of Missing Out) is infectious. The AI indicator is good as an independant indicator, just as is the EC indicator is. However those are both bull market constructs. So whilst I take note of buying pressure, the sum of my analysis continues to suggests that the markets are ripe for a black swan event and where one of the primary black swans that could explode is the banking sector, a chain reaction of debt deleveraging that results in the sale of liquid assets such as the tech giants. Of course the Fed will act to plug the hole with more money printing but as we saw in March 2020 it can still result in massive panic driven price movements.
Dow Stock Market Trend Forecasting Neural Nets
So far all attempts to predict the Dow via machine learning have failed, as mentioned above feeding nets with obvious data such as open, close, high, low and then a candle chart version does not work. In fact the neural net part is the easy part via the likes of Tensor Flow. The problem is with the data which as I explained earlier needs to be preprocessed i.e. feeding raw data into the networks as inputs ends up with noisy networks, if it did work then it would be easy to successfully train neural nets and they would be widespread instead as far as I am aware there aren't really any neural nets out there that can successfully trade stocks.
So what does preprocessing actually mean?
There are 2 ways to go about preprocessing data for stock market forecasts.
1. Pattern recognition
2. Trend Analysis preprocessing.
1. Pattern Recognition
One only needs to look at the deep fake videos out there of what actually needs to be done in terms of preprocessing of IMAGES, creating a dataset of hundreds of preprocessed Images of market price charts, say covering a fixed 60 data points (days) each with the goal of predict the next 20 or so days, and there is no guarantee it will work though appears to be the best route to go down, given the success rate of image recognition algorithms, though it would result in a huge neural nets for instance a chart image with a resolution of 256 by 256 pixels would require at least 65,636 input layers! i .e. as the images are broken up and fed to the network
So obviously most of the work, perhaps 95% is in generating the dataset of market chart images where 150 years of Dow data at 60 data points (days) per image stepped every 20 days translates into about 1500 images. Would that be enough? Maybe it could work for a trained / labeled neural net but likely not be enough data for deep learning, the holy grail of machine learning.
So the problem with forecasting markets using neural nets is not with the neural nets, that's the easy bit, but rather creating the dataset's, a lot of time consuming trial and error and I suspect I am going to have to model the neural nets on what I actually do when analysing markets i.e. looking it on varying time frames, zooming in and out.
Likely 60 day data set will not be enough, so require separate neural nets trained with 120 day, and 230 day (1 year) price chart images etc. That acts as preprocessed inputs to the next network, will that work?
Maybe I will need to add nuances by creating dataset's i.e instead of candle charts use swing charts, maybe point and figure etc. Maybe separate neural nets on each dataset that feeds its output as inputs into the next neural net.
2. Trend Analysis Preprocessing
Along with image recognition there is going down the EC route i.e. full spectrum trend analysis, which effectively would mean reinventing the wheel, writing thousands of lines of code in attempts to convert price data into trend analysis data points which basically means creating a whole host of of expert systems that pre-analyse the price data before it is fed into neural nets that along the lines with the EC indicator would each take a huge amount of fine turning, and likely end up with separate useful expert systems in their own right, I.e. trend analysis such as price patterns, support resistance, trend lines, MACD, seasonality, elliot waves. I say reinventing the wheel because I've already done all of this before and more from the Mid 1990's to the early 2000's when I created a multitude of market expert systems and automated technical trading tools, in fact I even had a primitive AI neural net attempting to LEARN from preprocessed data though of course one cannot compare then with now in terms of machine learning as we now have -
a. Infinitely greater compute power! There is no comparison between what one can do with a high end system today than one from 2000.
b. That using neural networks today's is a doddle! About a million times easier than in 2000! In 2000 one only had snippets of information i.e. layers of neurons all connected together by weights was about the extent of information available with much yet to be discovered / invented over the next 20 years.
So I have literally been there and DONE that over a period of about 7 years from around 1994 to 2001, creating what was probably at the time one of the worlds most sophisticated Trend Analysis development environments that I called "TA Dev", literally decades ahead of its time where I included everything I could think of during those 7 years such as Chart pattern recognition,, swing Trader, a primitive neural network, Elliott Waves generator,, seasonal analysis, trading systems galore, and the requisite full spectrum charting etc. But what it lacked at the time was PROCESSING POWER and probably also MEMORY to hold all of the data and variables so was very resources hungry.
A short 3 min video of running the program 20 years later. https://youtu.be/b68yZGc7AmA
The program was good for testing strategies and observations, conducting studies and pattern recognition but was overkill for what was needed to actually trade. During that time period I basically did TA to death! The program was written in Borland Delphi (pascal), which died a slow death during the early 2000's with the last iteration being Delphi 7 (2002) though the last version I actually used was Delphi 5. After Delphi 7 it all become a bit of a mish mash mess, a series of incompatible stop start nonsense that I will now need to sort through to see what could actually work as I attempt to resurrect 'Ta Dev" to use towards generating data for inputs to neural nets.
The last compiled executable of the trading development environment is dated 2nd October 2001 which I managed to get to run under Windows 10 under Window XP service pack 3 compatibility mode, but as has been the case for well over a decade the program fails to load any data into memory probably due to deprecated DLL's that the program is attempting to call.
So the first step is to get an Delphi IDE up and running and then get the code working so that it loads data. If this turns out to be impossible under Windows 10 then an XP system will need to be built.
The twin goals of the software will be to convert the various automated generators to provide preprocessed data for the neural net, and write new code to generate standardised charts for image pattern recognition, that should be straight forward to generate many hundreds of images of charts at the click of a button.
Crossing the Rubicon With These Three High Risk Tech Stocks
With Western tech corps trading on inflated bubbling valuations where is one to look for value? Phase 1 was to look at the downtrodden small cap biotech sector. Phase 2 was to look at the crypto's following the bursting of their bubble that has seen crypto's fall by 50% to 80% with much further to go.
So where next?
We'll there is a sector that has suffered a significant correction with stocks trading on relatively low multiples and unlike so called 'value' stocks have the potential to multiply many times. However, for this potential reward one carry's a far higher risk than one is used with the US tech giants.
The perfect stocks to buy would be our tech giants trading on much fairer multiples, but for that we are going to have to wait for the cookie to crumble with the less experienced investors and fund managers to realise that the likes of Nvidia are priced at 100 YEARS to earn it's share price!
So what do we do? We cross the Rubicon or the Yangtze river and seek value in that other worldly high risk high stakes poker game called CHINA!
So here follow 3 select CHINESE STOCKS that I expect to deliver between X3 to X4 over the next 5 years.
These 3 stocks are not something I've just come up with out of the blue by running screening exercises as was the case for the biotech stocks, but are well known large cap chinese stocks that have a track record that all of us have been aware of for some YEARS, in fact you might even be invested in some of them, though if you are hopefully you won't have bought them near their earlier highs as they have been on a bearish trend trajectory which gives my a window of opportunity that does not exist in the West to buy cheap when no one wants to touch them with a barge pole and sell them HIGH when everyone will be clambering to buy a stake at the highs just as is the case with the US tech stocks today.
BUT UNDERSTAND THIS, these stocks despite what their market capitalisation suggests ARE HIGH RISK! Both due to falling prey to the CCP and the US administration as we are literally crossing the Rubicon into enemy investing territory.
For instance there exists the risk of the stocks getting delisted which if it happens would make it a pain in the butt to trade out of them, However the simple solution to this would be to buy them on other exchanges such the Hong Kong, UK investors already incur a forex fee when buying US listings so buying in HKD is not that big of a deal, remember were not investing for few pennies, 10% 20%, but instead several hundreds of percent in which case the fx fees should not make much difference.
This could be a good time to accumulate a small position given that the chinese tech stocks are sharply lower year to date, down typically 20% to 30%, largely in response to increased scrutiny from the CCP that has been busy issuing billion dollar fines..
Baidu (BIDU) $172 - 0HL1B - HKG: 9888 - Google Clone
Baidu is a mini Google that continuous to keep a beady eye on what Google is upto so that it can then copy, ripping off what Google does by replicating it in China.
Now here's your electric shock - Google Trades in a PE of 35 / Market Cap of $1.78Trillion. Guess what Baidu trades on ?
Trailing P/e of 8.3, Market cap of $62 billion, now Google is a great corp but it's a lot easier to double by going from a market cap of $62 billion than $ 1.8 trillion ! And even if Baidu X10 from here it would still only be worth 1/3rd of what Google is today! Okay so forward P/E rises to 15 but that still puts the growth stock in the cheap zone (under a PE of 20).
In fact years ago I actually held shares in Baidu which I exited for my oft mentioned reasons of carrying geopolitical risks, so I know the stock well for many years. What has the stock done since I disinvested, first if was already weak going into Pandemic low, but then soared like a pocket rocket to new highs of $350 before giving up virtually all of the gains to currently stand at $173.
It's clearly in a downtrend but there is very heavy support under the market, from $150 to $200, and then at $140. So Baidu is virtually in the middle of it's major support 'buying' zone an area that is good to accumulate in, if only US stocks could come down to such accumulation zones.
I am tempted to buy an opening stake right now and then additional stake at around $154 that's about a 15% discount from the current price.
Objectives, well the initial objective is $340, it's recent high, beyond that depends on annual earnings growth, but the stock could easily find itself growing on an PE of 25+, so the stock is cheap in valuation terms. And I could easily see it trade to $750 in 5 years time.
So even though the trend is DOWN, I consider Bidu to be trading AT it's buying level of $173, with a secondary buying level of $154, so I will be buying a small stake in Baidu on Monday and maybe a similar amount again at around $154.
Alibaba - BABA $203 - 9988.HK - China's Amazon Clone
Alibaba trades on a trailing PE of 22 and forward PE of 23 with earnings growth of 22% and on a market cap of $565bn. The stock price is down 1/3rd on its Oct 2020 high of $320, currently standing at $203, trading AT significant support. at $200. The next support level lower is at $170 to $190 so I am not seeing much further downside.
Clearly the market does not like chinese stocks right now which could make this an opportune time to buy Alibaba shares on a PE of just 22, remember folks this is an Amazon clone, what does Amazon trade at ? A trailing PE of 68! More than TRIPLE Alibaba. So with little downside I am going to be buying exposure to this Amazon clone later today (Monday) and then again on any dip into the $190- $170 trading zone especially as apparently Cathy Woods ARK funds recently sold out of Alibaba and other chinese tech recently (after the price drop) which is a good buy signal :)
What about upside, extrapolating the earnings growth and say half Amazons PE than that would put the stock at X2.5 to X3.5 over the next 5 years, $530 to $730.
Tencent - TCEHY $69 - 0700.HK
Another Chinese Google Clone , internet services / AI etc. Over the past 5 years the stock X4 from a low of $25 to a high of $100, currently correcting lower by a significant 30% that puts the chinese tech giant on a market cap of $678bn and a PE of 25.5 which is a little high though Tencent does have decent earnings growth with a net profit margin of 25%.
So can the stock X4 again during the next 5 years?
Well despite the 30% correction to date the stock is still over priced, preference would be to look at buy at a PE of 20 or lower. The stock is currently trading in the support zone of $65 to $73 which if it fails to hold could send the price down to $53, which would be a 25% drop on the current price and translate into a PE of about 19. An invest at $55 would offer the scope to X3 or X4 over the next 5 years, therefore I will be seeking to buy a stake in Tencent at around a Buying Level of $55 that would put a target high during the next 5 years at between $165 and $220.
And remember folks the best time to buy stocks are when they are HATED and as the above charts illustrate chinese stocks are hated right now. I will be listing these stock along with further the high risk bio tech stock and see if I can generate an EC ratio and and ASVF price forecast in my next analysis.
CME Black Swan
I finally got around to converting one of my potential high probability black swan events into a video, the Earth being hit with a Massive Solar Coronal Mass Ejection the risks of which I shared with Patrons in my analysis of 24th November 2020.
I will take a look at the stock markets trend in a forthcoming analysis, whilst my next analysis will finally deliver the additional 5 biotech stocks with the potential to 10X and as a bonus I will be including a high risk tech stock also with the potential to 10X.
Your analyst who has a feeling that our beloved AI tech stock charts could soon start resembling those of their Chinese anti-matter counterparts.
Copyright © 2005-2021 Marketoracle.co.uk (Market Oracle Ltd). All rights reserved.
Disclaimer: The above is a matter of opinion provided for general information purposes only and is not intended as investment advice. Information and analysis above are derived from sources and utilising methods believed to be reliable, but we cannot accept responsibility for any trading losses you may incur as a result of this analysis. Individuals should consult with their personal financial advisors before engaging in any trading activities.