Market prediction based on macroeconomic indicators - page 38
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Thank you. I'll do some reading.
The hardest part of creating economic models is transforming the input data. If you look at economic indicators (there are about 10,000 of them), they differ from each other in many ways. Some grow exponentially, others twitch in some range, others twitch around zero with increasing magnitude, others change jerkily in the middle of history, etc. To create a model, all these data must be altered so that they have similar statistical characteristics that do not change over time. There are such possibilities:
1. Calculate the relative speeds: r[i] = x[i]/x[i-1]-1. This transformation automatically normalises the data, there is no looking into the future, you don't need to do anything else. But a big problem exists with zero data (x[i-1]=0) and negative data, and there are many of these in economic indicators.
2. Calculate increments d[i] = x[i] - x[i-1]. This transformation does not care about zero and negative data, but the increments grow over time for exponentially growing data such as annual gross product. I.e. the variance is not constant. For example, it is not possible to plot the dependence of GWP increments on the unemployment rate because the unemployment rate fluctuates within a range with constant variance, while GWP grows exponentially, with exponentially growing variance. So the increments must be normalized to the time-varying variance. But calculating the latter is not easy.
3. Remove from the data the trend calculated for example by Hodrick-Prescott filter and normalize the high-frequency residual by the time-varying variance and use as a model input. The problem here is that Hodrick-Prescott filter and other filters based on polynomial fitting(Savitzky-Golay filter, lowess, etc.) look ahead. Mooving lags the data and is unsuitable for trend removal, especially on exponentially increasing data.
Any other ideas?
There is a peek into the future in my last GWP growth prediction. I only discovered it after publication. That's why the model predicted past events so well. I keep struggling.
There is an opinion that when interest rates change, each time there is a scorrelation of existing dependencies and the formation of new ones... The rate has not changed in the last seven years....
the correlations between what?
correlation coefficients between macro factors cannot change globally
and the strong correlation between the variables in the model does not interfere?
So far, so good. So far it is difficult to find even one economic indicator capable of predicting the future value of GDP better than simply taking the latest known value of GDP. Actually, it is possible to find such a predictor, but only if one already knows this future value. For example, by looking at history, one can select good predictors and find a fairly accurate model of "past" future values. But this is self-defeating. At each point in history one must only take the data available until that moment, select predictors on that data and build a model of the future. But so far, the predictors chosen on past data are bad at predicting the future. "Poor" in the sense of worse than a trivial prediction equal to the last known value. Increasing the dimensionality of the model does not increase its accuracy. My theory is that if I can't find even one predictor, there is no point in looking for two predictors according to my numerous experiments. Replacing a linear model with a non-linear model only reduces the accuracy.
Converting input data by a second arbitrary creates a circular cloud on the plot <converted past future> -<converted predictor>, which confirms the absence of any correlation between the second derivatives. The predictions turn out to be almost equal to the trivial predictions. The correlation between the first derivatives exists, but is small. But again, the predictor chosen on the past is not suitable for predicting the future. Probably market crashes have different backgrounds. For example the crash of 2008 has different economic assumptions and predictors than the crash of 2001. Without derivatives, the data is very highly correlated and converting it to one range of variance is difficult without looking into the future. If there is time, I will explain this in more detail.
Here is an example of trivial predictions based on the use of the second derivative as a method for transforming the input data:
For those who want to try their methods of transforming inputs, selecting predictors and building a prognosticating model, I attach a Matlab data matrix. In the first column are the dates of the quarters. In the remaining columns are the different economic predictors. GDP is in the 1166 column (number of the first column with dates = 1). The date of the quarter is the first day of the quarter according to the Fed methodology, i.e. 1/1, 4/1, 7/1, 10/1. Quarterly data is assigned to the first day of the month of that quarter. For example, the most recent GDP value is ascribed to 10/1/2015, i.e. the beginning of the 4th quarter of 2015.
the correlations between what?
correlation coefficients between macro factors cannot change globally
There can be a change in the strength of the influence of macro data on the price of an instrument .......... i.e. the last seven years looked at non-farm ,
rate changed - look at wages and weekly benefit claims .
this is of course a different case and the time horizon is very different, but the correlation could be ..........
There may be a change in the strength of the influence of macro data on the price of an instrument .......... i.e. the last seven years have looked at non-farm ,
rate changes - look at wages and weekly benefit claims .
this is of course a different case and the time horizon is very different, but the correlation could be ..........
It is precisely the selection of model factors that is the challenge of the century. From a large array of economic factors, it is always possible to find those that give the best forecast on this data interval, but not the best on the other data interval. The problem is solved only by forward tests - not the best accuracy, but the same prediction accuracy for the training and forward samples.
After a lot of fiddling, I eventually came to the conclusion that the published macro forecast is better than mine and started using it.
http://library.hse.ru/e-resources/HSE_economic_journal/articles/18_01_07.pdf interesting article on the subject.....
There is such a fraudulent office in the field of economics that hands out paperwork called "Nobel Laureate". It's such a hangout, a huge hangout, a hangout covering up the financial bubbles of the west.
Look at any crisis. Did any of these Nobels or any of the "great" three predict these crises?
There's your answer.
Nobel must be given to the authors of "Down Game", which shows the whole underbelly of the stock quotes on the real estate market. It would seem that real estate, what could be more real than this reality? Nevertheless, the papers which displayed all these remarkable indicators invented by the Nobels were bogus, not real at all. People understood this, kept spinning....
And for an appetiser, watch "The Wolf of Wallstreet". The level of rubbish behind the "Nobel Laureate" signboard is in the details. THIS IS TRUE. And anything written by a bunch of Nobiles is a strategic lie. At best it can be used for short periods of time, like Granger.
On a larger scale, all that is written in the west on the level of macroeconomics is a game of three glasses. Because ECONOMICS is the PRODUCTION OF PRODUCTIONS AND SERVICES and FINANCE has NOTHING to do with ECONOMICS. This is a version of the children's game Monopoly for adults. Let everyone answer a simple question: let's take one of the most expensive companies in the world, Google. Tomorrow we throw it away, we destroy it. Will somebody starve to death, will they be left without trousers? We know all this from common practice. The sandpaper crisis of the early 2000s, when the index fell threefold. So what?
And finally.
Is the work of the author of the thread of interest? Undoubtedly. You have to eat every day. Very interesting work, but we must not forget that it relates to our everyday reality, and it must be understood that the Nobels in the field of market economics are crooks, nothing more. Nobels are nothing more than that. But taking a bunch of economic indicators, selecting from them those that affect the predicted indicator... That's very interesting...
Good luck to the author of the thread.
There is a fraudulent office in the field of economics which hands out paperwork called "Nobel Laureate". It's such a hangout, a huge hangout, a hangout covering up the financial bubbles of the west.
Look at any crisis. Did any of these Nobels or any of the "great" three predict these crises?
There's your answer.
Nobel must be given to the authors of "Down Game", which shows the whole underbelly of the stock quotes on the real estate market. It would seem that real estate, what could be more real than this reality? Nevertheless, the papers which displayed all these remarkable indicators invented by the Nobels were bogus, not real at all. People understood this, kept spinning....
And for an appetiser, watch "The Wolf of Wallstreet". The level of rubbish behind the "Nobel Laureate" signboard is in the details. THIS IS TRUE. And anything written by a bunch of Nobiles is a strategic lie. At best it can be used for short periods of time, like Granger.
On a larger scale, all that is written in the west on the level of macroeconomics is a game of three glasses. Because ECONOMICS is the PRODUCTION OF PRODUCTIONS AND SERVICES and FINANCE has NOTHING to do with ECONOMICS. This is a version of the children's game Monopoly for adults. Let everyone answer a simple question: let's take one of the most expensive companies in the world, Google. Tomorrow we throw it away, we destroy it. Will somebody starve to death, will they be left without trousers? We know all this from common practice. The sandpaper crisis of the early 2000s, when the index fell threefold. So what?
And finally.
Is the work of the author of the thread of interest? Undoubtedly. You have to eat every day. Very interesting work, but we must not forget that it relates to our everyday reality, and it must be understood that the Nobels in the field of market economics are crooks, nothing more. Nobels are nothing more than that. But taking a bunch of economic indicators, selecting from them those that affect the predicted indicator... That's very interesting...
Good luck to the author of the thread.
I agree that they give Nobels for crap, sometimes, but you can find a lot of interesting stuff there too ........... Have you seen the movie "margin call"?
Do you know what exactly a Nobel in economics is for?
Is there a single work or study on economics that deserves a Nobel and you have read and understood it completely?