Market prediction based on macroeconomic indicators - page 4

 
Demi:
There are too many indicators, factors and news that affect the price.

Exactly, too much.

Demi:

1. there is. Not for everything, but there is.

2. Why? The vast majority ofmacroeconomic indicators are forecasted and quite successfully by experienced analysts and put their forecasts out to the public a week or two in advance.

3. the market reacts to both the forecast and the possible deviation

Now let's analyze which is more important, the exchange rate or the economic indicators?

In other words, belief in money or money itself? The exchange rate is the belief in money, and money is just paper.

Let's assume that nothing happens in the economy, but the faith in money is undermined by noneconomic events, what can economic indicators predict?

But if you accept that the exchange rate is primary, then it is understandable why economic indicators change over time when the exchange rate changes.

If all of the above is true, then there is no point in making a forecast based on economic indicators because all waves and all influences have been taken into account by the price and the forecast can only be made by correctly decomposing the quotes into their components.

 
Urain:

Or another example is the money supply figures in the US.

Directly affect the stock market - they print dollars and inject them into the economy. Most of these dollars go to the stock exchange and inflate stock prices and the index accordingly.

Since some time, the Open Market Committee and the US Treasury have officially refused to publish these indicators in the public domain.

 
Urain:

Exactly, too much.

Now let's break down which is primary, the exchange rate or the economic indicators?

In other words, belief in money or money itself? The exchange rate is the belief in money, and money is just paper.

Let's assume that nothing happens in the economy, but the faith in money is undermined by noneconomic events, what can economic indicators predict?

But if we accept that the exchange rate is primary, then it is understandable why economic indicators change over time when the exchange rate changes.

If all of the above is true, then there is no need to make a forecast based on economic indicators because all waves and all influences have been taken into account by the price and the forecast can only be made by correctly decomposing the quotes into their components.

The relationship is two-way - so it is impossible to unambiguously determine what is the primary influence.

But simple models don't work here. As far as I have read, the Open Market Committee uses a simulation model of more than 100 equations to assess the market and make rate decisions in its day-to-day work.

 
gpwr:

...Predictive capacity can be measured by correlation coefficient or mutual information.

Can you elaborate on the mutual information. What is the method, how to use it? In which packages is it available?
 
Demi:

The connection is two-way - so you can't unequivocally determine which is the primary influence.

But simple models don't work here. As far as I read, the Open Market Committee uses a simulation model of more than 100 equations to estimate the market and make rate decisions in their day-to-day work.

OK, 100 equations, what data do they put in there? Current or forecast?

I assume current, otherwise we would have to believe in the integrity of presidents and the integrity of postmasters...

If they use current data, they will get the current state of the system, and then calculate the right regulators(interest rates, etc.) to get a stable development.

Just understand the purpose of regulating the economy, to stabilize it so the crisis only affects the part of the system that is ineffective. That is the intention, how they do it is another matter.

As a result, the indicators (regulators) are calculated to stabilize the economy, but it (regulation) is not always effective, and then the market comes to the rescue and adjusts the parameters that were not taken into account.

So you get that the market is just a system for decreasing errors in economic regulation. What follows from this is up to you.

 
Urain:

In sum, indicators (regulators) are calculated to stabilise the economy, but this (regulation) is not always effective, and then the market comes to the rescue and adjusts the parameters not taken into account.

So you get that the market is just a system for decreasing errors in economic regulation. You can deduce what follows from this.

How can the market adjust the unaccounted for parameters?

And how do you explain the occurrence of economic crises in this system then, if the market is a "system for reducing errors in the regulation of the economy"?

 
Demi:

How is it that the market can adjust unaccounted for parameters?

In this system, how do you explain economic crises, if the market is a "bug-reduction system"?

By changing the value of money and the public broadcasting of indices, that are used by many traders and economists alike.

ZS We trampled on Vasily here and he asked an interesting question:

C-42015.02.13 14:22EN
gpwr:

...Predictive ability can be measured by correlation coefficient or mutual information.

Can you elaborate on the mutual information. What is the method, how to use it? In which packages is it available?
 
C-4:
Can you elaborate on the mutual information? What is the method, how to use it? In which packages is it available?
In classification packages it is usually built inside. It's called inportance. See my article. Also available as separate package, e.g. Boruta, FSelector. All of them define some abstract value, which characterizes the "impact" or "importance" of the independent variable on the target variable.
 

I am not familiar with the terms and techniques you are discussing.

What if you select several types of data and by optimisation method assign a certain figure to each type of data and then output the sum of these figures?

And when the threshold of that amount is exceeded, open a trade.

 
C-4:
Can you elaborate on mutual information. What is the method, how to use it? In which packages is it available?

It's hard to explain in a nutshell. Read here first

https://ru.wikipedia.org/wiki/%D0%92%D0%B7%D0%B0%D0%B8%D0%BC%D0%BD%D0%B0%D1%8F_%D0%B8%D0%BD%D1%84%D0%BE%D1%80%D0%BC%D0%B0%D1%86%D0%B8%D1%8F

and then here (the chapter on mutual information, where the formula is given):

http://www.jclinbioinformatics.com/content/2/1/16