GOLD in a couple of weeks >1000 ? - page 6
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
I can't find the Dow Gold Ratio historical anywhere, can you tell me?
?????
"dow to gold ratio historic chart":
http://www.tradersnarrative.com/dow-jones-index-to-gold-ratio-historical-chart-983.html
http://home.earthlink.net/~intelligentbear/com-dow-au.htm
........................
?????
"dow to gold ratio historic chart":
http://www.tradersnarrative.com/dow-jones-index-to-gold-ratio-historical-chart-983.html
http://home.earthlink.net/~intelligentbear/com-dow-au.htm
........................
Thanks, but these are charts and need quotes. I took now the D1 yahoo data to DJI and GOLD for the last 5 years, calculated the ratio and tried to draw out the patterns and got this:
Interpretation: unfortunately the series is not linearly decomposable into its summation components. The next question is the predictability of the series, I will try to answer as quickly as possible.
Thanks, but these are charts and I need quotes. I have now taken the D1 yahoo data to DJI and GOLD over the last 5 years, calculated the ratio and tried to draw out the patterns and got this:
Interpretation: Unfortunately the series does not decompose linearly into its summand components. The next question is the predictability of the series, I will try to answer as quickly as possible.
The series is predicted with cof. Teil is 0.64 (the naive estimate is 1). Recommendation for October 30: do nothing....
Signal error: ratio of correctly guessed moves to the total number of moves.
Theil coef: Sum((predictedDirection - realDirection) * (predictedDirection - realDirection)/(naiveDirection - realDirection)*(naiveDirection - realDirection)) for each step, direction can be -1,0,1.
There is too much politics in the movement of the gold price. Gold is not a commodity, or rather not really a commodity.
Technical analysis is of little help here.
P.S. What kind of software is Specto 6.3, and even in Ukrainian?
There is too much politics in the movement of the price of gold. Gold is not a commodity, or rather not really a commodity.
TechAnalysis is of little help here.
P.S. And what for a program - Specto 6.3, and moreover in Ukrainian?
Specto is written at my firm for research purposes. The first step in time sequence research is to decompose the time sequence into orthogonal components which add up to the original series. Predicting the components and summing them up can be easier than predicting the original sequence. There should be materials on the orthogonalization method in the library of the Russian Academy of Sciences.
Note the following screenshot of Ratiotopia - it moves the window across the time series and makes a prediction in each step, thus validating the time series. The series is predicted!
Specto is written at my firm for research purposes. The first step in time sequence research is to decompose the time sequence into orthogonal components that add up to the original series. Predicting the components and summing them up can be easier than predicting the original sequence. There should be materials on the orthogonalization method in the library of the Russian Academy of Sciences.
Note on the screen of Ratiotopia - it moves the window around the time series and makes a prediction in each step, thus validating the time series. The series is predicted!
Well, I personally think this approach is quite reasonable (but hardly for gold, the market manipulation is higher there than for other "commodities"). Just which orthogonal components: sine, polynomial or what else?
Well, personally I think this approach is quite reasonable (but only hardly for gold, market manipulation is higher there than for other "commodities"). But which orthogonal components: sine, polynomial or what else?
The orthogonal components of geometric transformations method, they are linear in this case, are selected by RMS error criterion. Their projections are so called decomposition elements (also orthogonal, see fig. above), the sum of decomposition elements gives the original series.
They are, by definition, decorrelated (see corresponding dissertations in the Academy of Sciences of Ukraine and Russia). Here is a publication on the subject (NOT the time series http://www.nbuv.gov.ua/portal/natural/VNULP/Komp-nauky/2009_638/27.pdf).
The ort. components of the geometric transformation method, they are linear in this case, are distinguished by the mean square error criterion. They are, by definition, decorrelated (see corresponding dissertations in the Academy of Sciences of Ukraine and Russia ). Here is a publication on the subject (NOT time series http://www.nbuv.gov.ua/portal/natural/VNULP/Komp-nauky/2009_638/27.pdf).
WOW!!!
Oh, I mean, how cool.
(thoughtfully so) md-a-a... I used to write dissertations for Geetz, too. ....
WOW!!!
Oh, I mean, how cool.
(Mm-hmm. I used to write dissertations for Geetz. ....
:) in the figure you see orthogonal el. decompositions, their sum is the original series. I have corrected the description in the post above. There are time series where you can see very well e.g. cyclicities and main trend. I was hoping to get dow gold quotes, annual and mate to confirm the point of view of the wolves.