I Will Show You How To Predict The Stock Market With This Tool. PROOF!

I'm Sebastian James the stock market is predictable really really predictable what I'm about to reveal to you is groundbreaking it's brand new and unique to this channel it's based on a huge amount of personal research that I've been doing over the last few months so here goes this is the xbx over the last 95.2 years what is that well that is a graph for the S P 500 hmm and has a predictability of 96.1 percent wow okay this is new predictability what is that and why is it so high 96.1 percent that's incredible it's on the same level as you waking up tomorrow and they're still being air for you to breathe the predictability of a straight line is a hundred percent because it's a straight line I know what the next values are going to be random data no zero percent there's no predictability in random data at all the S P 500 is 96.1 which is right up there with a straight line that that's just amazing by the end of this video I will show you the proof of exactly how I arrived at this figure and for the first time perhaps ever you will see for yourself exactly how predictable the stock market really is okay let's define our terms what exactly do I mean by predictable well predictability measures given the amount of data that we have like the history of the stock market How likely are the future returns to be equal or very close to what we can predict in other words when I say predictability I mean really predictability in the classic sense that you would think about it but this is madness the share market goes up and down all the time every day it's oh it's up and it's down and then the share market crashes by 40 there's no way that the share market could be 96.1 percent predictable it just does not make sense and yet I'm telling you that's exactly what my original research shows but first let's see what else I can predict here is the xj0 over the last 30.3 years the xjo is the ASX 200 okay and its predictability is 87.1 percent huh well that's not as predictable as the S P 500 but yeah yeah still that's not bad here is the hasac over the last 20.7 years what's that that is a higher interest rate savings account with bonus interest cumulative in other words it's a return that you'd get if you stick your money in a bank account which has high interest on it and its predictability is 97.4 wow that is a brand new high so we find out that cash is actually more predictable than the stock market only just but it is predictably more predictable and it's good that our figures actually reflect that here's BHP so an individual stock over the last 35.1 years what is its predictability is 85.7 well I guess that makes sense an individual stock wouldn't be as predictable as the whole stock market so I've shown you a number of results but it's now time to find out exactly how I arrived at those figures here's the S P 500 over the last 95.2 years it's a graph I showed you earlier but wait there is something brand new what is that that dotted line in Orange that brand new line is in fact the line of best fit um okay so what's the line of best fit well it's all based on ice cream this is my ice cream truck apparently people do not want to buy ice creams in the middle of winter but on the hot summer's day I just can't keep up with demand what I decided to do is to take down some data the amount of sales I'm getting versus how hot it is here look at the results on the left hand side we have the sales in dollars on the x-axis we have the temperature which is in degrees celsius the blue dots are my actual results they've got a certain amount of variability in them but if I stick a line through the middle I can see that the blue dots actually follow that line fairly well the line of best fit so getting the raw data is easy I can tell you how much I sell in a day I can measure the temperature I can write these things down but how do I actually get this line of best fit this linear regression line well what I could do is I could just draw a bunch of lines right through my data and this is my stab at the Blue Line the red line the green line all possible lines are best fit now some of those lines are obviously much closer to the data than others the blue line is pretty pathetic the red line and the purple line yeah they don't actually look too bad really so when we talk about the line of best fit we're talking about straight lines only and what exactly do I mean by Best Fit if I plot on my data on a graph which is the black dots and then I do a line of best fit and we'll figure out how to do that in a moment then the distance the error between the predicted result which is the line of best fit and the actual results which is the black dot those red bars the error is in fact how we calculate that line of best fit so the liner best fit would be the one single line it's the best fit where those errors are minimized and that's how it works but the problem is if I take a whole lot of lines and draw them how can I find out exactly which is the best one because supposedly there is a best one can I take a classroom of 10 year olds get them to draw random lines and actually measure the errors and then over time experimentally work out which is the line of best fit yes you can if you can get your hands in a classroom of 10 year olds but other than that you can actually mathematically derive it and here's the formula which actually gives you the exact answer that formula we've covered in a previous video but let's look at our past graphs that I've shown you in this video and now let's add the line of best fit well here's the XJR over 30.3 years and aha it has the brand new line the line of best fit very good well it is going right through the middle I'd agree with that and I'm telling you that there's no better line that could possibly be the line of best fit than that one yes that's exactly what I'm telling you here's a hisac the bank account returns are in the blue and the red line is fairly close to it and I can kind of see that some of the results fall below the liner best fits are the sort of balance that outgo above the line of best fit and on average yeah it's as close as it possibly can be before I can believe that the stock market is actually 96.1 accurate which sounds highly unlikely to me I need to get a better sense of exactly how accurate this formula actually is let's try some sample data here are days down the left hand side from one to ten the stock market closed on those days one dollar on day one all the way down to ten dollars a perfect line it's predictable can you tell me what's going to happen on day 11 yes it's going to close 11 day 13 it will close at 13 100 predictable so let's plot that on a graph yes there we go so the liner best fit and the actual data are bang on top of each other just as you'd expect so the r squared of that is calculated to be a hundred percent r squared is a measure of predictability How likely is your data to fit the predicted line of best fit in fact we've been using r squared all along back in my graphing software so SBX the predictability is 96.1 percent that's the r squared value okay obviously the stock market does not grow at a predictable rate it's not a exactly perfect line so let's introduce some variability into our data there you go I've done it day four which should be for no it's suddenly become a five it's an error the r squared is 99 okay so it's not perfect but it's pretty close to it if we look at the actual graph aha I can see there's a line of best fit and I can see one of the days is actually outside that and you'll notice that the line of best fit is actually moved up a bit to try to compensate for it interesting a little error is good more must be better and so oh wow day two is becoming eight instead of a two day four day six and day eight they're all wrong totally wrong the r squared is 51 right okay so the data kind of fits out later best fitting kind of doesn't it's a bit wishy-washy and if we graph that yeah we'll see exactly that don't we there's a line of best fit which will always be a straight line so that part is not surprising but look how far the data is actually either side of it it's doing its best but it's yeah I wouldn't say it exactly Falls along that line of best fit what's the opposite of predictability Randomness so let me introduce that so here we go equals r-a-n-d that produces a random number down the left hand side we notice the closing figures are totally random 0.084 0.5894 random random random the r squared is zero which means it has no predictability at all if I graph that out there we go there's a line of best fit it's done its best but yeah it doesn't really match anything does it so far we found out that the liner best fit is actually derived mathematically so it's accurate and we found that r squared which is the level of predictability is also derived mathematically and shows how closely the data actually fits the liner best fit if I'm using the same formulas in my charting software is what I showed you in the spreadsheet which I am then also the figures in my charting software must be correct it's all the same math so there we go that 96.1 percent comes directly from linear regression and the r squared value that's exactly the math we just looked at to understand why our math actually tells us that the stock market is highly predictable and yet your own personal experience says the stock market is anything but it fluctuates all the time there's one more factor that we must factor in that is the factor of time itself this is the xbx or the S P 500 over one year apparently it's gone down and our line of best fit also goes down oh it's good to see it actually goes the other way of the one year the predictability is 20.8 percent that's basically closer to zero than it is a hundred yes so over one year the data shows the stock market is not particularly predictable at all let's pull up two years hmm actually I don't know if that's any better to be honest the predictability of that is 23.8 meaning that the predictability of that is Highly Questionable however if I go to five years haha I think I am noticing something slightly different the graph definitely goes up and down but I am actually noticing it's following that straight line the line of best fit interesting what's the predictability score it is 76 aha good so the more data I have then the more the data is actually following a line which is interesting so let me get more data his 10 years worth oh yes look at that it's definitely following that line of best fit there's no mistaking about that predictability is 93.8 wow that's jumped up the more data I've got the more predictable yeah so the reason why I had that grandiose figure at the beginning which is actually 100 real is because of the sheer amount of data I had it's almost 100 years worth so if I've got a high level of predictability up in the 90s does that mean I can predict the share price next week no it doesn't mean that at all but it does mean this given the huge amount of historical data we have and the line of best fit that we've drawn through that data there's a high probability that the future returns of the S P 500 will fall along that line of best fit now sometimes the stock market might do a bit better than the line of best fit other times worse but on average it will actually fairly closely follow the line of best fit and that's why we're up in the 90s some people say oh the stock market is so unpredictable I wouldn't put my money in there well sure over the short term it absolutely is as is borne out by a low predictability over the short term but if you step back a bit particularly when you get to decades no it becomes highly predictable the stock market does follow that line of best fit well this is good news if I pick my favorite index to invest in find the right ETF find out the right kaggar the compound annual growth rate that I'm particularly liking and actually multiply but by how many years I'm going to invest for what is the probability it'll get to roughly that right figure for example if I need eight hundred thousand dollars what is the probability it will get between say 750 and 850 000 well based on the data we have yeah it's pretty likely on my graph so far I've shown you the line of best fit but these student monks you will notice that my Straight liner best fit is not always straight and here we go the lineup is fit is definitely not straight for some reason it's curving upwards but you told me that the line of best fit had to be straight it's the linear regression linear means straight line if the stock market is in fact not linear it blows entire math right out of the water sadly this is true but I have a secret of my sleeve first we need to determine is the S P 500 the stock market actually linear or exponential so if it's gone up 200 points this year in 10 years time is it likely on average to go up 200 points as well to determine that first of all I need to understand what there's a difference between linear and exponential not everybody will know so let me give you an example here's a graph you'll notice there's a straight line and that's called linear and you'll notice there's a curved line called exponential right the exponential line has an interesting feature and that it is gross slowly over time and as time goes on it gets steeper and steeper That's a classic exponential curve now that we can positively identify what's linear a straight line and what's exponential a curved line let's look at the stock Mark and figure out exactly what it is oh you see that's a curved line it's going upwards clearly that's an exponential liner best fit well does that mean my math now goes all out the window because it only works on straight lines well let's take a look so day one is one day two I add one to it right so it goes one two three four down to ten okay we did this before r squared or the predictability value is a hundred percent as I'd expect however now I'm going to take the same spreadsheet instead of adding one I'm going to multiply it by 1.5 oh multiply that means exponential yes that's right so day one is still one day two is one point five and then I'll trust that the math is correct after that but it goes up by 1.5 every day and ends up at 38. let's have a look at the r squared how predictable is that well 81 percent um well it's not bad but it's definitely not 100 predictable is it let's graph that out oh I see the lineup best fit in blue of course is a straight line it can't be anything else but straight but the actual data which is in Orange is definitely exponential well that's a problem because regression Theory does not tend to work unexponential data so what do we do in the previous video one of my viewers said very interesting could you do the log of returns to get linear regression it would be nice to see both the r squared values and the slope of the best fit compared to the held beliefs on long-term kaggar the second part I'll leave for a future video but the idea of using log to straighten out exponential data is in fact very real the curvy line turned into a straight line on the left is in fact the bacteria over time which is an exponential curve and then we use a log function and sadly it becomes straight data that's nice one way of straightening out exponential data is to find a log of it but there is something called exponential regression I mean it's actually a thing take a look so if I Google it your exponential regression it's definitely a thing so we found it the secret to dealing with exponential data is to use exponential regression so exactly how do I do this step number one transform the data so that it allows for linear model what use a method of least squares to determine the linear model okay that standard linear regression that we did before step three transport the data and the model back to the original form in other words use exponential functions oh no so the way that exponential regression works is take your data transform it into linear data by using a log function now that everything is linear we can just go ahead and get our blind the best fit exactly as we've always been doing and then once you've got that you take those values and you turn them back into exponential data so while exponential regression sounds good in reality it doesn't exist it's all linear regression and you're just transferring two linear and then back to exponential that to me sounds like a hack it's about time exponential regression got its own formula rather than relying on linear regression but that's the way it works the question is does this series of steps which are stitched together with duct tape actually work well let's have a look in column two the close column here is the exponential data we had before with the r squared of 81 percent which really didn't work for me very well so I'm going to take that data and I'm going to pop that now in a different spreadsheet but it's exactly the same data it's exponential I'm now going to take the natural log of it that's the Ln part so the natural log of 1 is 0 the natural log of 1.5 is 0.40 and so on down the list I'm now going to take those numbers which have the log function put on them and transfer them back to our original table so day one is zero day two is 0.405 okay so they're the log values transferred and haha the r squared which is the predictability is suddenly become a hundred percent wow how did that happen so it would appear that if you take exponential data and you actually take the log of it and then pass through your linear regressions formula your line of best fit that we love and have been doing this entire video it does actually a hundred percent work mind blowing and here's a graph to prove that this is a linear model and the actual data falls on the line that best fit exactly right now how do I go back to the exponential data well there's a formula it's got x's and y's in it and if I graph that suddenly we find the exponential data actually fits the line of best fit wow it all lines up amazing now I can't really see much because like the line of best fit is exactly on the data and therefore I can't actually see any difference so let me introduce a small error so we can see a difference into day five I'm going to introduce a 20. no that's wrong it should be five point something and yet the error is now 20. if I graph that out oh look at that in Orange I can find out that that value is in fact wrong and you notice that the line of best fit which is actually an exponential line of best fit you could tell that because of the way it's curved up the whole line is actually pushed up a little bit now of course exponential lines of best fit come from linear ones so we can graph it in the linear and yes there we go the error which came from the exponential data is clearly shown in the linear data now because I can put it in linear I can do an r squared and find out the predictability number it's not going to be 100 but what is it well r squared is 88 ah okay not bad 88 does match with what I'm visually seeing in that graph the question is in my charting software which I've written for the ground up myself what formula am I actually using I'm actually using linear regression or exponential regression the answer is both which we can see on the graph yes that's definitely curving up it's definitely exponential liner best fit but because you're smarter than the average bear you now know that exponential regression actually relies on linear regression underneath the hood so now when I throw up figures like the S P 500 over 95.2 years is in fact 96.1 percent predictable you know the figure is mathematically correct because I've taken you through the entire process of how we derive that what it shows is the real Return of the stock market very closely matches the line of best fit which we calculated and it does that for two reasons number one the stock market is fairly predictable no matter how much data I've got if the stock market was random we would not get a hundred percent would get something closer to zero so there is inherent predictability in the stock market and a lot of it plus I have huge amounts of data almost 100 years worth of data and the formula has worked out with that amount of data and with how predictable it is it is very confident that the stock market future returns will closely match that liner best fit and therefore it's highly predictable so now that I have a brand new superpower what in the future can I do with it when we look at individual stocks or individual ETFs and they have a particular return sometimes it's only like 10 years worth of data how confident am I that that data is actually going to be reflected into the future in other words if I invest in vhy for example how reliable are the returns going to be well now we've got the math we can actually say categorically whether that data is good you know up in the high 80s or 90s and therefore it might be something worth considering or if it's not is it down in the 30s and 20s and therefore regardless of how high the return actually is I don't believe it can actually achieve that now this actually lives in my charting software that means in future videos we'll be able to use this predictability figure again and again and again now all this is 100 original research I've never ever seen this applied to shares before now of course the math is real it's just linear regression and exponential regression but as far as using this to predict how reliable the stock market returns are no this is all original research now I've stepped you quickly through all the formulas but what if you want to spend a bit more time looking into each of those formulas to see exactly how this regression Works firsthand click here to find out about that or if you've seen that click here [Music]