This is remarkably interesting.
Thank you very much.
I will Attempt to implement this within my dictionary
B.
This is remarkably interesting.
Thank you very much.
I will Attempt to implement this within my dictionary
B.
6 Structure (s)I'm responding to your post, as an excuse to re-post....people tend to only read the final page of a thread only. If I re-post it will be on a fresh page and hopefully viewed longer.Originally Posted by ;
Serial Correlation....
The deficiency of daily serial correlation in market markets is frequently the starting point for random market concept. Studies as early as Labys and Granger (1970) confirm the daily randomness. Even in trending markets, the expectation of upward price or down price for the following day is still about 50-50. Even odds in non-trending markets isn't hard to comprehend; in trending types it is not so obvious. A market that's trending up or down will possess an enhanced probability from the fashion direction; it is simply pretty nicely masked by the random element that's always present. By way of instance, a tendency of one cent per day (on average) is practically lost in the average daily trading range of, say, ten cents. By egory, nevertheless, the random element can be averaged from the underlying price movement could be seen. We'll illue this point.
Assessing one day's information to the previous days information (or comparing one candlestick into the next candlestick, or comparing one price to the next price) following the market this manner, are examples of linear steps.
Now if you follow the information, linearly, day to day in sequential/chronological order, the market appears to move randomly. There is no day to day signs of a underlying trend.
This is the point where the random theory conspiracist's view this as some type of evidence that the markets are a random walk. When you take a look at the information in a linear, chronological, sequential, string (time series) it merely appears to be random This demones is that even in a fashion there is no day to day serial correlation. Even in trending markets, the expectation of upward price or down price for the following day is still about 50-50
instead of linearly stick to the information price to price. Lets begin measuring the exact same exact information, using non linear steps used in market analysis. Here we'll use a two day sample of information. As opposed to following the information day to day in sequential order. The results look just as random as following the market daily. There is still no signs of the underlying downtrend Even with a two day sample of information.
Let us look at a three day sample of information.... Now things begin to get interesting..Same exact data. . Still using non linear measures.Now you begin seeing some signs of the underlying downtrend. By using a three day sample of information you started to filter from the day to day random changes (random element )
It also becomes more apparent with the 4 day sample of information.
What exactly does this prove? .... Do the mathematics
Testing a Trend for Serial Correlation
There is one long downward trend from March 6, 1991, through July 10, 1991. This conduct covered 88 trading days (127 calendar days) or just over four weeks. The entire price fall is 121 cents or 1.375 cents per trading day. The daily trading-range average over the 88 trading days is 8.17 cents, which range between a high of 29.25 plus a low of 3.25 cents. Table 1-5 has the prices along with a list of higher/lower prices on the basis of 1 day, two days,... through 5 times.
Considering that a downtrend has been Pre-selected, we've added a bias to the downside.
Therefore ties those cases in which the compared prices were that the same-will be given to the H, or higher count. That contributes to Table 1-6
One of those basic fundamentals in AMVT is an understanding that the markets are not linear. Employing linear steps to explain analyze a non linear market is similar to using a tape measure to measure how many gallons it takes to fill a bucket
Should you set your information, group prices over the correct sample of period (price over time) rather than look at the information linearly, day to day through a string of candlesticks, you find the markets are in no way a random walk.
A market that's trending up or down will have an enhanced probability from the trend management
To examine that statement in a downtrend using the data beginning with the 1 day sample of information.
If you traded in the course of the last day's tendency (following price) with the probability of trading with the underlying trend, you would have correctly traded in the direction of the underlying tendency 52% of the time and you would have been incorrect 48% of the time. Correctly trading in the direction of the underlying trend following price day-to-day is still pretty much a coin toss.
Let us look at the two day sample.had you traded in the direction of the underlying trend following the two day samples of information rather than following price, you would have correctly traded in the management of the underlying tendency 62 percent of the time and you would have been incorrect 38% of the time.
Now things get interesting. .
Let us look at the three day sample.had you traded in the direction of the underlying trend using the three day samples of information rather than following price, You would have correctly traded in the direction of the underlying trend, 70 percent of the period and incorrect 30% of their time.
4 evening sample you would have been correct 73 percent of their time and incorrect 27% of the time.
5 day sample you would have been correct 76 percent of their time and wrong 24% of the time.etc.
This is what you call proof of principle...
Market profile is built on the assumption of comparing the current day's profile into previous day's profile. MP methodology is built on the assumption that the market is some the day to day serially connected
I've only demoned and proved that there is no day daily serial correlation in the market. It doesn't matter if you're following price day daily or following daily statements, the results are still the same. Even in trending markets, the expectation of upward price or down price for the next day is still about 50-50
MP methodology is built on the premise/assumption that the day daily daily profiles are serially correlated. We only proved in the previous article that the market has no day to day successive correlation..When examined, the evidence doesn't support MP methodology. Comparing yesterdays profile into the current day's profile in an effort to trade in the direction of underlying tendency, the expectation of upward price or down price for the next day is still about 50-50. You would have similar odds throwing darts in the chart.
So when I go on and on about this not being a MP thread..it's since we dont trade working with a series of daily profiles, we dont examine today's daily profile with yesterday's daily profile, since the evidence demones there isn't any day to day serial correlation in the market. .
Hey mzvega
I've some thoughts regarding your prior writeup about serial correlation. Very interesting BTW.
I attempted to implement the procedure in my calculation of flow.
I compared the latest 1D profile and it has mention points to a profile (and it is reff points) from three days past
I did a crude backtest to view how the method fails during distinct states (trending, bracketing). I found that through prolonged trending conditions the calculation of direction flow matches fairly well (if DIR was positive, market moved up).
However as soon as the market was bracketing for 5 days or more, the DIR flow calculation stopped up on the opposite side of where the market proceeded later (if DIR was positive, market had a breakout into the down side).
Can you use this methodology or did you post to prove a point about traditional MP concept?
Anyhow I truly enjoy whatever you have written and I look forward to your future articles.
Sir mzvega,
If I'm allowed to diverge a little, this jumping rather than serial correlation seems to tie up using DeMarks method of comparing not consecutive but a few pubs back.
Hope this does not disturb the flow of thread.
Due
The methodology in the way you described above?
No more...I follow the directions in the source material for net flow analysis.
I don't understand exactly what you mean by Dir and you believe this Dir is used as some kind of signal to trade. What is the purpose of all those web flow calculations in the spreadsheet if you don't plan on utilizing the data for analysis?
Before I spend my time answering any questions regarding net flow or some other spreadsheet calculations, how can you copy paste the source material that you used to understand the goal of using net flow analysis, what it calculates, and how the data is used for analysis before using that data to construct a spreadsheet.
Can you also copy paste among the examples used in the source material that shows how to browse and use the data.
This way I know my time wont be wasted trying to help explain something to someone who hasn't invested their own time studying.
Then we could compare notes....what you believe your calculating compared to what the source material says how and exactly what you should be calculating and much more important how to use the data. .
There is not any skipping per se, this isn't something that's applicable to a bar chart or something comparable to TA.Sir, If I am allowed to diverge somewhat, this jumping rather than serial correlation appears to tie up using DeMarks system of comparing not consecutive but a few bars back. Hope this will not disturb the flow of thread. Thanks
I asked that others read VBPT..before squandering their time on trying to compute a spreadsheet. I keep to explain that you can not put the cart ahead of the horse.
That sequential correlation article has been posted to this thread a few times.
It also is loed in the very first chapter of VBPT...
it's most likely the first most important basic fundamental needed to learn AMVT. It is the most significant basic building block for analysis and producing spreadsheets for analysis. It give you a starting point to get a spreadsheet. Since you learn about bracket display, the market unit, VA rule....etc. Then you understand context as you expand refine your own dictionary. As your knowledge base expands your spreadsheet will expand to include other things you're able to read and understand. As you read and learn you discover that references imply different things in different contexts, you then organized your spreadsheet into a readable format.the depth of your spreadsheet should reflect the depth of your understanding.
I find it fruing that everyone has created spreadsheets without learning the basic fundamentals needed to really use it.
People who really was able to read this first chapter of VBPT would have known it had been used to establish a point about traditional MP concept...
I posted it to establish a point that no one reads the thread....or the source material....not even the first few pages of VBPT
I completely disagree, the argument and sources are so broad and dispersive that your constant reference to not reading the sources is depressing and intimidates who approaches the subject. Me, like many others, we have certainly read the resources, and certainly not entirely understood, even due to the time spent studying the subject. So something that's relevant, at first reading might appear irrelevant.
Building a database and spreadsheet is a tool to understand how things operate and also to examine them, not a place the cart before the horse.
And asking for clarifiions online forum a part of learning procedure.
Having no longer access to the website with free weekend analysis the only way would be to face each other with our calculations and also our databases by correcting the errors of interpretation. Nobody is obliged to answer but please, in precisely the exact same time, maybe not treat others as beggars.
Have not been trading often for a while but I have manage to steal some transactions while still in the workplace.
EURUSD happen to be bracketing for 16 days because the gap up. 10 and 15 days are bracketing, 3 days trending up; flow is up with TFF show growth in speed( lower is better). Trades are only place just prior to UK opening to the 10 day bracket upper limit.
GBPCHF was best capping since 4-5 days past under enormous volume. Reversing down bringing with is a flow for 3 days. TFF enhance last 2 days also. A market was put on 5 days Lower Limit.
USDCAD similar to the GBPCHF above. It was capping 7-8 days back under heavy volume. Trajectory down to last 3 days with a surge in rate yesterday as TFF value shrinks. 5, 3, 10 days bracketing with commerce loion under their lower limits. Trade is still developing...
Posting Permissions
Reply With Quote
