Quick Links You last visited August 21, 2017, 5:21 pm All times shown are Eastern Time (GMT5:00)  Mathematics and the LotteryCan a winning lottery system be created with existing math formulas? United States Member #124493 March 14, 2012 7023 Posts Offline
 Posted: July 13, 2012, 5:36 pm  IP Logged  
Hi RL/DrSan, Both approaches look interesting, but my focus is elsewhere right now. Of the articles I posted, the ones on bursty behavior will probably get my attention first. I see potential in the probabilities they can produce and that it will give some direction on which are the best numbers to bet. So, I will defer any comments on what you have both posted because I simply haven't done my homework enough to feel like I can give you a quality reply. RL Thanks again for posting your program. The Double Odd and Double Even filters are simply outstanding. A member some time back posted an intriguing idea about repeatability. Basically he said that out of the group of numbers picked in the big games, that reliable reduction percentages can be produced by comparing the amount of numbers that can and will repeat over differents windows in time. Another thought for you is...have you tried applying your filters to other filters? Try a last time picked history, for example, and apply the same concepts you've used on the historical picks. That is currently something that I am plugging away at now. I beleive that applying good reduction techniques to all my subfilters is as important as applying them to the numbers picked. Anyway, best of luck to the both of you and I'm sorry I couldn't respond intelligently to your ideas. I really like this burstiness concept...and is something that i have tracked to get wins... as i sit here in front of my excel spreadsheet i am trying to anticipate what numbers will burst, and why? What groups will burst...how do burst look and what are the best bursts... I am not a programmer so i do most of my analysis the old fashioned way... i understand filters to a point...but the best filter is my brain. I have simplified my pick 3 and pick 4 approach to anticipating repeaters as the game can be daunting if you try too much... any way JKing...good luck. Recently in pick 4...three of the four numbers repeated...in a burst.... burstiness happens in a approximate cycle of 36 games...   
bgonÃ§alves Brasil Member #92564 June 9, 2010 2264 Posts Online  Posted: July 13, 2012, 6:14 pm  IP Logged  
I really like this burstiness concept...and is something that i have tracked to get wins... as i sit here in front of my excel spreadsheet i am trying to anticipate what numbers will burst, and why? What groups will burst...how do burst look and what are the best bursts... I am not a programmer so i do most of my analysis the old fashioned way... i understand filters to a point...but the best filter is my brain. I have simplified my pick 3 and pick 4 approach to anticipating repeaters as the game can be daunting if you try too much... any way JKing...good luck. Recently in pick 4...three of the four numbers repeated...in a burst.... burstiness happens in a approximate cycle of 36 games... Hello, boner lotto, a lottery for 39/5 matrix in the previous post is 100% in any lottery Catch always 1st step to separate the first digit 13 and the last digit 09 These formations from 0000 to 3333 can take all such pairs = 0022 or all odd 1113, taking it further diminishes the 5th random number is 30 to 39 the number 56 Example 38 may give evidence with which training from 0000 to 3333 is not out yet But lottoboner the 37 groups are 100%, but attention is half the sitema, missing endings ok   
bgonÃ§alves Brasil Member #92564 June 9, 2010 2264 Posts Online  Posted: July 13, 2012, 6:59 pm  IP Logged  
Hello lootoboner, fantasy 5 california test = 6823 = 15 24 27 29 31 5 digit number the first 1_2_2_2_ leatorio 3039 gave 31 1,2,2,2 Lottoboner these are within the 37 groups because it is 100% Sure miss the last digit, 5,4,7,9 5th number is random, does not enter the study In Example 31, the biggest problem is terminaçaoes, but always one or two repeats Endings of the previous test. then we have the 37 groups of 27 or 31 to 39 The accuracy is 100%, but only four endings that you can use like a pick 4 That forum has a lot of filter systems and implement   
United States Member #124493 March 14, 2012 7023 Posts Offline
 Posted: July 13, 2012, 7:04 pm  IP Logged  
Hello lootoboner, fantasy 5 california test = 6823 = 15 24 27 29 31 5 digit number the first 1_2_2_2_ leatorio 3039 gave 31 1,2,2,2 Lottoboner these are within the 37 groups because it is 100% Sure miss the last digit, 5,4,7,9 5th number is random, does not enter the study In Example 31, the biggest problem is terminaçaoes, but always one or two repeats Endings of the previous test. then we have the 37 groups of 27 or 31 to 39 The accuracy is 100%, but only four endings that you can use like a pick 4 That forum has a lot of filter systems and implement so what is your interesting bursty structure?   
bgonÃ§alves Brasil Member #92564 June 9, 2010 2264 Posts Online  Posted: July 13, 2012, 7:18 pm  IP Logged  
Hello, lottoner, you have to use 37 as a fixed or reference, ode to choose, as having frequencies of 37 Average hot and cold, to reduce, or example the number 38 = see which formations From 0000 to 3333 who had not left with the number 38, to reduce even more the terminations You can play all pairs = 2,2,4,8 or all odd = 1137, the 5th random number is ok, the four endings, you can see by the statistics in each vertical position, or as a pick4! Of the 37 that are 100% not escape, are perfect because there is only one formation per drawing. So practically 5 fantansy was like a pick 4   
United States Member #124493 March 14, 2012 7023 Posts Offline
 Posted: July 13, 2012, 8:03 pm  IP Logged  
Hello, lottoner, you have to use 37 as a fixed or reference, ode to choose, as having frequencies of 37 Average hot and cold, to reduce, or example the number 38 = see which formations From 0000 to 3333 who had not left with the number 38, to reduce even more the terminations You can play all pairs = 2,2,4,8 or all odd = 1137, the 5th random number is ok, the four endings, you can see by the statistics in each vertical position, or as a pick4! Of the 37 that are 100% not escape, are perfect because there is only one formation per drawing. So practically 5 fantansy was like a pick 4 when i say structure i am referring to the pictures of which the algorithm would apply...this is from the video posted plus the bursty nature... are you saying all odd forms some sort of interesting structure?   
bgonÃ§alves Brasil Member #92564 June 9, 2010 2264 Posts Online  Posted: July 13, 2012, 8:08 pm  IP Logged  
Boolean operator In Matlab, there are four logical (boolean) operators: Boolean operator: Meaning: & Logical and  Logical or ~ Logical NOT (addon) xor Exclusive or These operators produce vectors or matrices of the same size as the operands, with 1 when the condition is true, and 0 when the condition is false. Given x = array [0 7 3 5] and y = array [7 8 2 0], these are some possible operations: Operation: Result: n = x & yn = [1 0 0 1] = ~ m (y  x) = m [0 0 0 0] xor = p (x, y) p = [1 0 0 1] Once the output logical Boolean operations or is a vector or matrix with only 0 or 1, the output can be used as the index of an array to extract the appropriate elements. For example, to see the elements of x which satisfy both conditions (x < y) and (x < 4), you can write x ((x < y) and (x < 4)). Operation: Result: x < y ans = [1 1 1 0] x < 4 ans = [0 1 0 0] q = x ((x < y) and (x < 4)) q = [0 3] In addition to these Boolean operators, there are several useful builtin logical functions, such as: any True if any element of a vector is true all True if all elements of a vector are true exist true if the argument exists isempty true for an empty array isinf true for all infinite elements of an array isNaN true for all elements of an array that aranumber find find indexes of nonzero elements of an array Relational operators There are six relational operators in Matlab: Relational operator: Meaning: < less than <= less than or equal > greater than > = greater than or equal == equal (no assignment) ~ = not equal These operations result in a vector to array the same size as the operands, with 1 when the relationship is true, and 0 when it is false. Data Arrays x = [3 5 0 7] and y = [0 2 8 7], these are some possible relational operations: Operation: Result: k = x < y k = [1 1 1 0] k = x y = k < = [1 1 1 0] k = x = y = k = [0 0 0 0] Although these operations are typically used in conditional commands, such as ifelse statements to branch to different cases, they can be used to make very complex array manipulation. For example x = y (y > 0.45) finds all the elements in the vector y such that y i > 0.45 and stores them in the vector x. These operations can be combined with Boolean operators, too.   
bgonÃ§alves Brasil Member #92564 June 9, 2010 2264 Posts Online  Posted: July 13, 2012, 9:20 pm  IP Logged  
  
United States Member #59354 March 13, 2008 4322 Posts Offline  Posted: July 14, 2012, 12:51 am  IP Logged  
This is an interesting idea RL...but what bursty pattern structures are you looking for?? Take a win four game for example...a bursty pattern contains large structural elements of repeateability so to speak... these bursts occur in intervals... how will 0 0=0 help predict and anticipate burstiness?? LB I got a little off topic and the logicgates are not part of the bursty patterns. Prediction is not one of my strong points and I am not sure if it can be done in any manner that will consitantly hit. Even a broken clock is correct twice each day meaning that even if something shows good it may be just chance unless it can be repeated many times within a short time frame. I believe that if using probability to come to some sort of prediction then your results will fall very close to the expected. In many of the analysis that I run I find clusters of hits and misses and from this data I find that I can recognise when a data will start a cluster and approxmently when it will end. This does not mean that I can make an exact prediction but rather I know when to start playing some value. The problem is finding enough clusters from different data sets that all converge at or near the same time. Bursty analysis may prove very useful in this but will require much work to find a solution. RL   
United States Member #124493 March 14, 2012 7023 Posts Offline
 Posted: July 20, 2012, 3:23 am  IP Logged  
7/10  18  7  0  1  13  1  8  0  22  10  15  12  10  1  16  2  17  21  5  4  4  1  8  5  0  4  8  6  3  5  0  11  2  4  9  0  1  9  16  7/11  19  8  1  0  14  2  9  1  23  11  16  13  11  2  17  0  18  22  0  5  5  2  0  6  1  5  9  7  4  6  1  0  3  5  10  1  2  10  17  7/12  20  9  2  1  0  3  10  2  24  12  0  14  12  3  18  1  19  23  1  6  6  3  1  7  0  0  10  8  5  7  2  0  4  6  11  2  3  11  18  7/13  21  10  3  2  1  4  11  3  0  13  1  15  0  4  19  2  20  0  0  7  7  4  2  8  1  1  11  9  6  8  3  1  5  7  0  3  4  12  19  7/14  22  11  4  3  2  5  12  4  1  14  2  16  1  5  20  3  0  1  1  0  8  5  3  0  2  2  12  10  7  0  4  2  6  8  1  4  0  13  20  7/15  23  12  5  0  3  0  13  5  2  15  3  17  2  6  21  4  1  2  2  1  9  6  0  1  3  3  13  0  8  1  5  3  7  9  2  0  1  14  21  7/16  0  13  6  1  4  1  14  6  0  16  4  18  3  7  0  5  2  0  3  2  10  7  1  2  4  4  14  0  9  2  6  4  8  10  3  1  2  15  22  7/17  1  14  7  0  5  2  15  7  1  17  5  0  4  8  1  6  3  0  4  3  11  8  0  3  5  5  15  1  10  3  7  5  9  11  4  0  3  16  23  7/18  2  15  8  1  6  3  0  8  0  18  6  1  5  9  2  7  4  1  5  4  0  9  1  4  6  6  16  2  11  4  8  6  10  0  5  1  4  17  0  7/19  3  0  9  2  7  0  1  9  1  19  7  2  6  10  3  8  5  0  0  5  0  10  2  5  7  7  17  3  12  5  9  7  11  1  6  2  5  18  1  7/20  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39 
  
Economy class Belgium Member #123700 February 27, 2012 4035 Posts Offline  Posted: August 2, 2012, 8:25 pm  IP Logged  
07/20/2012  0309133537  07/19/2012  0206181921 
  
United States Member #124493 March 14, 2012 7023 Posts Offline
 Posted: August 12, 2012, 6:47 pm  IP Logged  
I apologize to all players regarding my last post... I did not specify the game i am playing... but i posted to show examples of "burstiness"...the flag formation could be a bursty behaviour.... All pink highlights are hot as in hot pink formatting... the chart is a Matrix Walk aka DSH... i forget which game...either 5/39 or 6/59, i also forgot the date, although the chart indicates the date... Bursty!!!!!!!   
bgonÃ§alves Brasil Member #92564 June 9, 2010 2264 Posts Online  Posted: September 7, 2012, 7:08 pm  IP Logged  
Considering the main issues of the regression models in the econometric analysis, in practice, the researcher often faced with a situation where the regression is given a "bad", ie, tstatistics, most estimates are small, reflecting the insignificance of relevant independent variables. At the same time, the F statistic can be quite large, which indicates the significance of the regression as a whole. One possible reason for this phenomenon is known as multicollinearity and occurs when there is a high correlation between the factors. One of the conditions of the classical regression model is the assumption of linear independence of the explanatory variables. When that condition, ie when one of the variables is a linear combination of the others, is called a complete collinearity. In this situation, you can not use ordinary least squares (OLS). In practice, the complete collinearity occurs very rarely. Much more often faced with a situation where among the factors that there is a high correlation. Then indicate the presence of multicollinearity. In this case, the least squares estimation (least squares method) formally exists but has a "bad" properties. Multicollinearity can occur due to various reasons. For example, several independent variable may have a tendency common time, which undergo small oscillations. In particular, this can happen when the value of an independent variable values are dating each other. only some of the most characteristic signs of multicollinearity. One. A small change in the source data (such as the addition of new observations) leads to a significant change in the estimates of the model. 2nd. The estimates have large standard errors, low value, while the overall model is significant (high determination coefficient R2 and the corresponding Fstatistics). 3rd. Coefficient estimates are incorrect from the point of view of the theory of signs or oversized. do you do when all signs that there multicollinearity? The answer to this question is no, and among econometricians have different opinions about it. When confronted with the problem of multicollinearity can be a natural desire to discard the "extras" independent variables, which can serve as its cause. However, be aware that it can cause new problems. First, it is not always clear which variables are redundant in this sense. Multicollinearity means only approximately linear relationship between the factors, but not always allocate the "extra" variables. Secondly, in many cases, removal of all independent variables can have a significant impact on the essence of the model. Finally, the rejection of variables called essential, i.e., truly independent variables that affect the dependent variable studied, resulting in a change of the model. In practice, usually when a multicollinearity factor cleaned at least for the analysis and then repeat the calculations   
Thread Starter United States Member #5599 July 13, 2004 1196 Posts Offline  Posted: September 10, 2012, 11:58 pm  IP Logged  
Fyi, Programming languages for free. http://numpy.scipy.org/ http://code.google.com/p/mpmath/ You are a slave to the choices you have made. jk Even a blind squirrel will occasionally find an acorn. There is no elevator to success, you will have to take the stairs.   
United States Member #116268 September 7, 2011 20244 Posts Offline
 Posted: September 25, 2012, 3:58 pm  IP Logged  
Simple math, is all that is needed.   
