Hearst index - page 44
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
specialised statistical packages, namely the statistical analysis package R.
As a working option I can also recommend Stata. Plus, actually, for writing a paper, it is quite kosher to learn TeX. As for the rest, I completely agree with C-4.
Good afternoon! In general, the idea was as follows: calculate the H indicator, build a function on the price of metal, then impose the change in the cost of production of this metal and analyse the change (the so-called profit that can be obtained by the organisation). Analyse in terms of those factors that are influenced and not influenced by external factors.
I, frankly, intuitively understand that this is little like something worthwhile, because, well, it should be programmed, such skills do not have. But the supervisor of my dissertation strongly recommended to use this factor in calculations. So it turns out to be nonsense.... at the initial stage. ((((
A few more words, now about gold.
You write about gold in terms of fundamental analysis. Obviously, the most important relationship for you will be the price of gold with its cost. But that relationship has yet to be found and proven. Markets are not stationary, and if a relationship is understood as a simple correlation, then it is not enough for the price series. There is also cointegration - causality tests, for example, from this area, but there are still requirements for data normality, and price data does not have that. So, a very innovative approach for you would be if you first developed a non-parametric model for finding these very nonlinear relationships based on fractal statistics, and then tested this model on the relationship between gold production and its value. There is a complete vacuum in this field right now and there is a lot to write about. We may start from a simple assumption that spread between two linked series will not be a white noise, and if say, two persistent series, give antipersistent spread, it would indicate that there is the mutual state correction between these series (spread tend to zero, while being non-stationary), and therefore there is a relationship between the two series, although any other methods will show that there is no relationship, because there is no stationarity and no linearity!
Another piece of data that you should examine for correlation is the information on the hedges of the gold producers. There is a lot of data in the article for fundamental analysis.
I can also recommend Stata as a working option. Plus, actually, for writing work, it is very kosher to learn TeX. Otherwise, I completely agree with C-4.
Thank you so much !!!!! Off to explore!!! )))
Thanks so much !!!!! Off to explore!!! )))
If you decide to take up R, - I can recommend this book to you:
To be honest, the book is pretty clueless, but the only one of its kind, so you don't have to choose.
one of a kind
Apart from that in Russian.
Apart from that in Russian.