| Author |
Message |
     stoian
Member
Username: stoian
Post Number: 146
Registered: 03-2004Rating: N/A
Votes: 0
|
| | Wednesday, September 07, 2005 - 10:43 pm: |  
|
Betlong,
Where are you? I have received your message and i have answered you many times. I am ready for your test. Establishing the Market Profile for a title, means already TO FORSEE=TO PREDICT THE FUTURE EVOLUTION, ONLY IF YOU ARE KNOWING AND UNDERSTANDING VERY GOOD THE HISTORY OF THIS EVOLUTION. Please, look for to my forecasts done by me in the past year or even this year, until this moments: EUR/USD, Oil, FTSE 100, and also for the forecast made today for STP.AX.
Regards,
Doru Stoian
|
betlong
Member
Username: betlong
Post Number: 14
Registered: 07-2005Rating: N/A
Votes: 0
|
| | Thursday, September 08, 2005 - 12:43 am: |
|
Doru,
I haven't seen any replies to the last post at http://forum.incrediblecharts.com/messages/6/535552.html, nor received any email replies.
Before I start preparing the datasets, can you please say exactly what will be forecast, as otherwise the results of the test may be very unclear.
I've had a look at the STP post, and I'm finding it difficult to interpret. It may help greatly if you could walk through the specifics of what the chart is forecasting, in simple terms.
Subscribe to our FREE
monthly updates
Subscribe to regular Incredible Charts
updates, with new articles on Trading
Strategy, Trading Psychology, Money
Management, Tricks of the Trade,
New Indicators and the Economy.
|
|
|
|
|
stoian
Member
Username: stoian
Post Number: 150
Registered: 03-2004Rating: N/A
Votes: 0
|
| | Thursday, September 08, 2005 - 03:16 am: |
|
betlong,
In very simple terms: to know what will be happen tomorrow or after tomorrow, in other words for the "very nearest future" what means also - with a very small estimation error - the present time, or the short time future, each investor/speculator is needing AT FIRST a forecast for a medium to long term. The simple reason for is that: WE NEED TO DEFINE (TO PROOF AND PROVE AT FIRST=BEFORE ALL, IF THE CONDITION OF EXISTENCE FOR THE PICTURE IS FULFILLED)or in other words with the same meaning: we need to define AT FIRST THE WHOLE PICTURE, AND ONLY AFTER THAT DEFINITION WE CAN LOOK FOR ANY OF THE DETAILS EXISTING ON THIS PICTURE.
To define the picture, means in our concrete case: TO KNOW AT THE BEST THE HISTORY, which means and represents THE FRAME OR THE MAT FOR THE PICTURE.
If isn`t any history, than it gives not any present or any future. It seems simple enough for you the explanations until here?
When yes, then you will rapidly understand, that ON OUR PICTURES CAN BE MANY DETAILS, AND MOST OF THEM CAN BE VERY, VERY IMPORTANT, BUT OTHER DETAILS NOT.
We can also recognize under the very important details of the picture=chart, that such details like: "TREND IS YOUR FRIEND TILL THE BEND AT THE END" are very, very significant for us, if we are able TO CORRECTLY IDENTIFYING THE TREND AND TO SEE SO THE POSSIBLE BEND.
We have here the concrete case for STP.AX. We can see very clear THE BEND AT THE END OF THE DECREASE PERIOD. Much more clear is this image about the BEND TIME AT THE END OF THE DECREASE PERIOD for STP.AX, when we transform the scale for the price from a normal one to a logarithmical one, so that the fourth degree polynomial curve, used by us AS THE BEST FIT APPROXIMATION and represented on a normal scale, is not interrupted, but on the logarithmical scale APPEARS AS INTERRUPTED. That means in the very simple mathematical language: A GAP FOR THE DECREASING EVOLUTION MANIFESTED BY THE PRICE IN TIME, WAS ACTING ONLY FOR A WHILE, AND AFTER THE GAP WAS COMING TO HIS END=IS GONE, THAN THE PRICE WILL INCREASE AGAIN.
To be short with my explanations, i will say also that a forecast for short periods of time - from a day to another, from a week to another, from a month to another, are mostly very hard to be done, BECAUSE OF THE VERY=HEAVY RANDOM CHARACTER DESCRIBED BY SOME EVOLUTIONS, or THE ALSO CALLED FACTOR OF UNCERTAINTY - ASSUMED, MEASURED AND INCLUDED IN OUR FORECAST METHOD - can have an intensity level that CAN BE OUT OF THE NORMAL LIMITS.
So, you will become an answer from me for your possible questions, like:
1) What will be the very probable future evolution for the given series of time, for the title called X?
2) What are - most exactly=accurate - THE VERY PROBABLE LIMITS FOR THE SUPPORT AND RESISTANCE, or also called SIGNALS TO BUY, respective TO SELL?
If you need to know more aspects as i told you until yet, please name them to me.
Regards,
Doru Stoian
|
betlong
Member
Username: betlong
Post Number: 16
Registered: 07-2005Rating: N/A
Votes: 0
|
| | Thursday, September 08, 2005 - 06:40 pm: |
|
Doru,
I'm afraid I'm still very unclear about what is intended to be forecast with your technique.
Perhaps some specific forecasts with regard to STP - since you've recently analyzed it - may help. For example, are you able to say:
- is the black line in your STP diagram meant to be a forecast of the future STP price?
- if so, what probability would you say this has?
- if not, what is the meaning of the black line?
- what are the features in your STP diagram that identify things like:
*support and resistance levels,
*price targets,
*likely timeframe for achieving those targets
*any other features that you may think should be
identified as useful?
|
stoian
Member
Username: stoian
Post Number: 153
Registered: 03-2004Rating: N/A
Votes: 0
|
| | Friday, September 09, 2005 - 12:17 am: |
|
betlong,
The black curve in my STP diagram shows us: THE VERY PROBABLE PATTERN OR FORM (because of the level of the r-squared indicator which is telling us that a fourth degree polynomial were THE BEST FIT for the empirical historical individual evolution) FOR THE NEXT=FUTURE EVOLUTION FOR THE STATISTICAL CHARACTERISTIC CALLED THE MEAN OF THE PRICE ORIENTED=DICTATED BY TIME (you are also living oriented in time? Or not?)
Because this statistical characteristic is also an object of variation, exactly like in case of the empirical individual values, who are depending on the direction and amplitude of this variation, i should draw a second black curve on the chart, parallel and simultaneously with the first one, and also the rest of the features you have mentioned, but for me it was so clear like daylight (because of THE SUPPORT LEVEL FOR THE PRICE, who WAS ALREADY NOT ONLY EXCEEDED BUT ALSO EXPIRED IN TIME) that the MEAN OF THE PRICE will continue to rise in time so much, that i have thought will be useless to draw on the chart the other features mentioned by you, like:
*support and resistance levels;
*price targets;
*likely timeframe for achieving those targets;
*any other features that you may think should be
identified as useful.
But for your test, i will certainly show you all this features, and after that we will follow together if my forecast becomes true indeed.
Regards,
Doru Stoian
|
betlong
Member
Username: betlong
Post Number: 17
Registered: 07-2005Rating: N/A
Votes: 0
|
| | Wednesday, September 14, 2005 - 01:25 am: |
|
Doru,
I haven't forgotten about the forecast test - it's just that I've been caught up with more pressing matters.
Will get back with the data on the weekend sometime.
|
betlong
Member
Username: betlong
Post Number: 18
Registered: 07-2005Rating: N/A
Votes: 0
|
| | Saturday, September 17, 2005 - 07:55 pm: |
|
Doru,
Here's a dataset for a first go at a forecasting test. From my recollection, the dataset complies with the criteria discussed in previous posts:
- the data is from the security's original listing date, and contains more than 1500 data points.
- the dataset is up to a past, arbitrary date. The complete dataset will be posted once the forecast is published.
- the dates have been replaced by a sequential number.
Whatever the results come out to be, they won't be conclusive, with just one sample. Hopefully however the meaning and significance of these forecasts will become clearer. Good luck.
|
stoian
Member
Username: stoian
Post Number: 175
Registered: 03-2004Rating: N/A
Votes: 0
|
| | Tuesday, September 20, 2005 - 09:28 pm: |
|
Betlong,
Thank you for the data set, that you have sent me.
Why do you have replaced the calendar dates with sequential numbers?
Do you want to play the villain with me again?
You don`t understand really nothing about statistics?
You don`t really know about that thing, that to proof and to establish a REAL statistical correlation=link between the evolutions in time for two variables, which from the second type of variable - in our concrete case, the daily quotes for the security you will need the forecast - is more or less correlated=depending from the evolution of a first type of variable, in our concrete case THE TIME?
Why do you have not given me THE REAL NUMERICAL VALUES of this series of time AND HAVE REPLACED THIS REAL NUMBERS WITH ON YOUR WISH (OR EVEN RANDOM SELECTED) CHOSEN SEQUENTIAL NUMBERS?
I have told you already the very important aspect: THAT THE ENGLISH PROVERB "TIME IS MONEY" IS NOT QUIT A JOKE AND I WILL EXPLAIN YOU THIS AGAIN IN VERY FEW WORDS.
Therefore, if you need to buy or to sell a security with/against money, YOU WILL NEED ALSO TO KNOW - FIRST OF ALL - HOW EXPENSIVE OR HOW CHEAP IS YOUR MONEY AT THE SAME TIME YOU/THE MARKET IS EVALUATING THROUGH A QUOTATION THE QUANTITY OF YOUR MONEY THAT YOU WILL BRING IN EXCHANGE ON THE MARKET FOR THE RESPECTIVE/CHOSEN SECURITY.
So, because i am also using the settings "offered" by Microsoft ONLY=EXCLUSIVELY for Romania, on my SCATTERING CHARTS WILL APPEAR THE SO CALLED VARIABLE OF INFLUENCE "THE TIME" WHICH IS IN FACT THE SO CALLED VARIABLE OF INFLUENCE "THE MONEY", THAT WILL HAVE ONLY THE REAL VALUE FOR THE MONEY VALID ONLY FOR ROMANIA`S CURRENCY MARKET, OR ELSE SAID BUT WITH THE SAME ESSENCE: WHAT IS WORTHING - EXPRESSED IN ROMANIAN NEW LEU (AN IMAGINARY=AN EXOTIC CURRENCY OF AN EXOTIC COUNTRY) - A UNIT OF THE FOREIGN EXCHANGE BASKET (A REAL CURRENCY).
Once i am knowing THAT because i am smart enough: WHAT IS THE ROMANIAN MONEY WORTHING FINDING THIS THROUGH THE MOST ACCURATE VALUE OF YOUR REAL CURRENCY EXPRESSED IN ROMANIAN MEW LEU,I CAN PROOF THE CORRECT CORRELATION BETWEEN THE EVOLUTION OF YOUR REAL MONEY AND THE EVOLUTION FOR THE PRICES=QUOTATIONS FOR THE SECURITY YOU NEED THE FORECAST.
I am so sorry to say to you, that i not beleve anymore you mean very serious that you need to know what forecasts are indeed.
It seems to me early, you want like many other, to laugh about my work and to make her useless before the profane eyes of the most forumers.
So, please forget my promise!
Regards,
Doru Stoian
|
jimdene
Member
Username: jimdene
Post Number: 13
Registered: 07-2005Rating: N/A
Votes: 0
|
| | Wednesday, September 21, 2005 - 12:13 am: |
|
I feel like the little boy who said, "But the Emperor has no clothes on ", amongst such eloquent and utter incomprehensible language, but does it matter what happens in the future, don't we have to trade as we see it now and make decisions when the time comes
|
mjs12
Member
Username: mjs12
Post Number: 3
Registered: 06-2005Rating: N/A
Votes: 0
|
| | Tuesday, September 20, 2005 - 11:27 pm: |
|
Doru,
I'm also very interested to see the outcome of your forecast.
I don't understand why specific dates matter. Why don't you take Betlong's data and set the date of the most recent quote exactly 1 year ago.
I have done it for you, please see the attached file. I have adjusted the dates to account for weekends too.
Kind Regards,
Matt.
|
betlong
Member
Username: betlong
Post Number: 19
Registered: 07-2005Rating: N/A
Votes: 0
|
| | Wednesday, September 21, 2005 - 01:22 am: |
|
Doru,
I have no interest in laughing at your work. As I said at the very start: I take a skeptical, but open minded approach. Until there's evidence available, I try to not accept or dismiss any method, including yours.
I thought that when we agreed to do this exercise, it was a good way to cut through the confusion of the language barrier and other matters, and hopefully get an indication of the usefulness or otherwise of your forecasting method.
I don't understand what your issue with the dataset is. It was agreed that the dates would be shifted by an arbitrary amount, so that discovery of the security's identity would be impractical. Replacing sequential dates with sequential numbers is an easy way of doing this. In any case, it is an easy operation in Excel to replace those numbers with sequential dates - or you could have asked me to do it.
Your choice of not proceeding is ok with me, but please don't do this behind a smokescreen of accusations.
|
stoian
Member
Username: stoian
Post Number: 177
Registered: 03-2004Rating: N/A
Votes: 0
|
| | Wednesday, September 21, 2005 - 08:01 pm: |
|
BETLONG,
To better understand that i am not acuse you about nothing, maybe only about - however, a little ignorance - because it seems for me you don`t want to SIMPLIFY the way shown by me to become a better trader with the help of forecasts for you and also for other forumers, but you are only continuously trying - LIKE THE CHURCHES AND THE FREEMASONRY - TO MAKE ANY KIND OF LIFE IMPOSSIBLE=UNACCOUNTABLE=INCOMPREHENSIVE.
Do you really understand that thing: IF I WILL FOLLOW YOUR ADVICE=CONSULTATION=IMPOSSIBLE CONDITION FOR DOING MY FORECASTS, THE RESULTS WILL BE WRONG BECAUSE OF A VERY WRONG APPROACH YOU HAVE PROPOSED TO ME? The expected BEND AT THE END WILL NOT APPEAR "AT THE RIGHT TIME" IF I AM USING WRONG=IMAGINARY=UNREAL VALUES FOR THE VARIABLE CALLED TIME WHICH IS IN FACT ANOTHER VARIABLE CALLED MONEY, AND THAT BOTH OF THIS VARIABLES, THE MONEY INFLUENCING THE PRICES=QUOTATIONS SHOULD BE REPRESENTED ON TIME! If i follow your advice, the real values for the variable money shown in time, will be when not missing than will be certainly wrong=not the most appropriate for the evolution ALSO IN TIME of the second variable=the market price=quotation of a security.
Please, for a better understanding about what i mean, read my post also about on the thread called: La Profezia! The da Vinci`s last unbroken code?
Look on the site of Mr. Robert Langdon for the original document of da Vinci called La Profezia posted on the site, and you will see that his secret is starting from three rectangles symbolising THE HOLY TRINITY=THE FIRST HOLY FAMILY=THE FATHER, THE SON AND THE HOLY GHOST, and try to understand that for making PROPHECIES=FORECASTS you have to start from the three most important types of errors mentioned by me, WHICH LIKE BY THE HOLY FAMILY ARE TOGHETHER IN FACT A SINGLE ONE TYPE OF ERROR=THAT WHEN TRYING TO ESTIMATE THE PLACE OF THE REAL=TRUE MEAN FOR ANY PHAENOMENA YOU ARE RESEARCHING FOR.
But please, again, read my post on the thread La Profezia, and you will understand ALL the problematic of FORECASTS.
Regards,
Doru Stoian
|
stoian
Member
| |
Test Dataset
Data with dates added