Hi

I would very much appreciate some help writing what should be some simple PCFs for V7. I have spent a lot of time looking as past threads on this and the results to the PCFs are hit and miss. I am not getting anywhere at the moment the below attempts do not work.

Below is a list of the PCFs i need and where i have got to with them. All are standard 12,26,9 settings.

A. Is Monthly Macd >= the signal line.

(XAVGC12.21 - XAVGC26.21) >= (XAVG(XAVGC9,12.21) - XAVG(XAVGC26.21,9))

B. Is Weekly Macd >= 0 line.

(XAVGC12.5 - XAVGC26.5) >= 0

C. Is Weekly Macd >= the signal line.

(XAVGC12.5 - XAVGC26.5) >= (XAVG(XAVGC9,12.5) - XAVG(XAVGC26.5,9))

Thank you in advance.

Chris

Understanding MACD
Cascades of Moving Averages

Please try the following Condition Formula instead:

.03489263 * (C + 25 / 27 * (C21 + 25 / 27 * (C42 + 25 / 27 * (C63 + 25 / 27 * (C84 + 25 / 27 * (C105 + 25 / 27 * (C126 + 25 / 27 * (C147 + 25 / 27 * (C168 + 25 / 27 * (C189 + 25 / 27 * (C210 + 25 / 27 * (C231 + 25 / 27 * (C252 + 25 / 27 * (C273 + 25 / 27 * (C294 + 25 / 27 * (C315 + 25 / 27 * (C336 + 25 / 27 * (C357 + 25 / 27 * (C378 + 25 / 27 * (C399 + 25 / 27 * (C420 + 25 / 27 * (C441 + 25 / 27 * (C462 + 25 / 27 * (C483 + 25 / 27 * (C504 + 25 / 27 * (C525 + 25 / 27 * (C546 + 25 / 27 * (C567 + 25 / 27 * (C588 + 25 / 27 * (C609 + 25 / 27 * (C630 + 25 / 27 * (C651 + 25 / 27 * (C672 + 25 / 27 * (C693 + 25 / 27 * (C714 + 25 / 27 * (C735 + 25 / 27 * (C756 + 25 / 27 * (C777 + 25 / 27 * (C798 + 25 / 27 * (C819 + 25 / 27 * (C840 + 25 / 27 * (C861 + 25 / 27 * (C882 + 25 / 27 * (C903 + 25 / 27 * (C924 + 25 / 27 * (C945 + 25 / 27 * (C966 + 25 / 27 * (C987 + 25 / 27 * (C1008 + 25 / 27 * (C1029 + 25 / 27 * (C1050 + 25 / 27 * (C1071 + 25 / 27 * (C1092 + 25 / 27 * (C1113 + 25 / 27 * (C1134 + 25 / 27 * (C1155 + 25 / 27 * (C1176 + 25 / 27 * (C1197 + 25 / 27 * (C1218 + 25 / 27 * (C1239 + 25 / 27 * (C1260 + 25 / 27 * (C1281 + 25 / 27 * (C1302 + 25 / 27 * (C1323 + 25 / 27 * (C1344 + 25 / 27 * (C1365 + 25 / 27 * (C1386 + 25 / 27 * (C1407 + 25 / 27 * (C1428 + 25 / 27 * (C1449 + 25 / 27 * (C1470 + 25 / 27 * (C1491 + 25 / 27 * (C1512 + 25 / 27 * (C1533 + 25 / 27 * (C1554 + 25 / 27 * (C1575 + 25 / 27 * (C1596 + 25 / 27 * (C1617 + 25 / 27 * (C1638 + 25 / 27 * (C1659 + 25 / 27 * (C1680 + 25 / 27 * (C1701 + 25 / 27 * (C1722 + 25 / 27 * (C1743 + 25 / 27 * (C1764 + 25 / 27 * (C1785 + 25 / 27 * (C1806 + 25 / 27 * (C1827 + 25 / 27 * (C1848 + 25 / 27 * (C1869)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))) + .4396511 * (C + .8 * (C21 + .8 * (C42 + .8 * (C63 + .8 * (C84 + .8 * (C105 + .8 * (C126 + .8 * (C147 + .8 * (C168 + .8 * (C189 + .8 * (C210 + .8 * (C231 + .8 * (C252 + .8 * (C273 + .8 * (C294 + .8 * (C315 + .8 * (C336 + .8 * (C357 + .8 * (C378 + .8 * (C399 + .8 * (C420 + .8 * (C441 + .8 * (C462 + .8 * (C483 + .8 * (C504 + .8 * (C525 + .8 * (C546 + .8 * (C567 + .8 * (C588 + .8 * (C609 + .8 * (C630))))))))))))))))))))))))))))))) >= .41062481 * (C1 + 11 / 13 * (C21 + 11 / 13 * (C42 + 11 / 13 * (C63 + 11 / 13 * (C84 + 11 / 13 * (C105 + 11 / 13 * (C126 + 11 / 13 * (C147 + 11 / 13 * (C168 + 11 / 13 * (C189 + 11 / 13 * (C210 + 11 / 13 * (C231 + 11 / 13 * (C252 + 11 / 13 * (C273 + 11 / 13 * (C294 + 11 / 13 * (C315 + 11 / 13 * (C336 + 11 / 13 * (C357 + 11 / 13 * (C378 + 11 / 13 * (C399 + 11 / 13 * (C420 + 11 / 13 * (C441 + 11 / 13 * (C462 + 11 / 13 * (C483 + 11 / 13 * (C504 + 11 / 13 * (C525 + 11 / 13 * (C546 + 11 / 13 * (C567 + 11 / 13 * (C588 + 11 / 13 * (C609 + 11 / 13 * (C630 + 11 / 13 * (C651 + 11 / 13 * (C672 + 11 / 13 * (C693 + 11 / 13 * (C714 + 11 / 13 * (C735 + 11 / 13 * (C756 + 11 / 13 * (C777 + 11 / 13 * (C798 + 11 / 13 * (C819 + 11 / 13 * (C840 + 11 / 13 * (C861))))))))))))))))))))))))))))))))))))))))))

Plese try the following Condition Formula instead:

.1539843 * (C + 11 / 13 * (C5 + 11 / 13 * (C10 + 11 / 13 * (C15 + 11 / 13 * (C20 + 11 / 13 * (C25 + 11 / 13 * (C30 + 11 / 13 * (C35 + 11 / 13 * (C40 + 11 / 13 * (C45 + 11 / 13 * (C50 + 11 / 13 * (C55 + 11 / 13 * (C60 + 11 / 13 * (C65 + 11 / 13 * (C70 + 11 / 13 * (C75 + 11 / 13 * (C80 + 11 / 13 * (C85 + 11 / 13 * (C90 + 11 / 13 * (C95 + 11 / 13 * (C100 + 11 / 13 * (C105 + 11 / 13 * (C110 + 11 / 13 * (C115 + 11 / 13 * (C120 + 11 / 13 * (C125 + 11 / 13 * (C130 + 11 / 13 * (C135 + 11 / 13 * (C140 + 11 / 13 * (C145 + 11 / 13 * (C150 + 11 / 13 * (C155 + 11 / 13 * (C160 + 11 / 13 * (C165 + 11 / 13 * (C170 + 11 / 13 * (C175 + 11 / 13 * (C180 + 11 / 13 * (C185 + 11 / 13 * (C190 + 11 / 13 * (C195 + 11 / 13 * (C200 + 11 / 13 * (C205)))))))))))))))))))))))))))))))))))))))))) >= .07414684 * (C + 25 / 27 * (C5 + 25 / 27 * (C10 + 25 / 27 * (C15 + 25 / 27 * (C20 + 25 / 27 * (C25 + 25 / 27 * (C30 + 25 / 27 * (C35 + 25 / 27 * (C40 + 25 / 27 * (C45 + 25 / 27 * (C50 + 25 / 27 * (C55 + 25 / 27 * (C60 + 25 / 27 * (C65 + 25 / 27 * (C70 + 25 / 27 * (C75 + 25 / 27 * (C80 + 25 / 27 * (C85 + 25 / 27 * (C90 + 25 / 27 * (C95 + 25 / 27 * (C100 + 25 / 27 * (C105 + 25 / 27 * (C110 + 25 / 27 * (C115 + 25 / 27 * (C120 + 25 / 27 * (C125 + 25 / 27 * (C130 + 25 / 27 * (C135 + 25 / 27 * (C140 + 25 / 27 * (C145 + 25 / 27 * (C150 + 25 / 27 * (C155 + 25 / 27 * (C160 + 25 / 27 * (C165 + 25 / 27 * (C170 + 25 / 27 * (C175 + 25 / 27 * (C180 + 25 / 27 * (C185 + 25 / 27 * (C190 + 25 / 27 * (C195 + 25 / 27 * (C200 + 25 / 27 * (C205 + 25 / 27 * (C210 + 25 / 27 * (C215 + 25 / 27 * (C220 + 25 / 27 * (C225 + 25 / 27 * (C230 + 25 / 27 * (C235 + 25 / 27 * (C240 + 25 / 27 * (C245 + 25 / 27 * (C250 + 25 / 27 * (C255 + 25 / 27 * (C260 + 25 / 27 * (C265 + 25 / 27 * (C270 + 25 / 27 * (C275 + 25 / 27 * (C280 + 25 / 27 * (C285 + 25 / 27 * (C290 + 25 / 27 * (C295 + 25 / 27 * (C300 + 25 / 27 * (C305 + 25 / 27 * (C310 + 25 / 27 * (C315 + 25 / 27 * (C320 + 25 / 27 * (C325 + 25 / 27 * (C330 + 25 / 27 * (C335 + 25 / 27 * (C340 + 25 / 27 * (C345 + 25 / 27 * (C350 + 25 / 27 * (C355 + 25 / 27 * (C360 + 25 / 27 * (C365 + 25 / 27 * (C370 + 25 / 27 * (C375 + 25 / 27 * (C380 + 25 / 27 * (C385 + 25 / 27 * (C390 + 25 / 27 * (C395 + 25 / 27 * (C400 + 25 / 27 * (C405 + 25 / 27 * (C410 + 25 / 27 * (C415 + 25 / 27 * (C420 + 25 / 27 * (C425 + 25 / 27 * (C430 + 25 / 27 * (C435 + 25 / 27 * (C440 + 25 / 27 * (C445))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))

Please try the following Condition Formula instead:

.03489263 * (C + 25 / 27 * (C5 + 25 / 27 * (C10 + 25 / 27 * (C15 + 25 / 27 * (C20 + 25 / 27 * (C25 + 25 / 27 * (C30 + 25 / 27 * (C35 + 25 / 27 * (C40 + 25 / 27 * (C45 + 25 / 27 * (C50 + 25 / 27 * (C55 + 25 / 27 * (C60 + 25 / 27 * (C65 + 25 / 27 * (C70 + 25 / 27 * (C75 + 25 / 27 * (C80 + 25 / 27 * (C85 + 25 / 27 * (C90 + 25 / 27 * (C95 + 25 / 27 * (C100 + 25 / 27 * (C105 + 25 / 27 * (C110 + 25 / 27 * (C115 + 25 / 27 * (C120 + 25 / 27 * (C125 + 25 / 27 * (C130 + 25 / 27 * (C135 + 25 / 27 * (C140 + 25 / 27 * (C145 + 25 / 27 * (C150 + 25 / 27 * (C155 + 25 / 27 * (C160 + 25 / 27 * (C165 + 25 / 27 * (C170 + 25 / 27 * (C175 + 25 / 27 * (C180 + 25 / 27 * (C185 + 25 / 27 * (C190 + 25 / 27 * (C195 + 25 / 27 * (C200 + 25 / 27 * (C205 + 25 / 27 * (C210 + 25 / 27 * (C215 + 25 / 27 * (C220 + 25 / 27 * (C225 + 25 / 27 * (C230 + 25 / 27 * (C235 + 25 / 27 * (C240 + 25 / 27 * (C245 + 25 / 27 * (C250 + 25 / 27 * (C255 + 25 / 27 * (C260 + 25 / 27 * (C265 + 25 / 27 * (C270 + 25 / 27 * (C275 + 25 / 27 * (C280 + 25 / 27 * (C285 + 25 / 27 * (C290 + 25 / 27 * (C295 + 25 / 27 * (C300 + 25 / 27 * (C305 + 25 / 27 * (C310 + 25 / 27 * (C315 + 25 / 27 * (C320 + 25 / 27 * (C325 + 25 / 27 * (C330 + 25 / 27 * (C335 + 25 / 27 * (C340 + 25 / 27 * (C345 + 25 / 27 * (C350 + 25 / 27 * (C355 + 25 / 27 * (C360 + 25 / 27 * (C365 + 25 / 27 * (C370 + 25 / 27 * (C375 + 25 / 27 * (C380 + 25 / 27 * (C385 + 25 / 27 * (C390 + 25 / 27 * (C395 + 25 / 27 * (C400 + 25 / 27 * (C405 + 25 / 27 * (C410 + 25 / 27 * (C415 + 25 / 27 * (C420 + 25 / 27 * (C425 + 25 / 27 * (C430 + 25 / 27 * (C435 + 25 / 27 * (C440 + 25 / 27 * (C445)))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))))) + .4396511 * (C + .8 * (C5 + .8 * (C10 + .8 * (C15 + .8 * (C20 + .8 * (C25 + .8 * (C30 + .8 * (C35 + .8 * (C40 + .8 * (C45 + .8 * (C50 + .8 * (C55 + .8 * (C60 + .8 * (C65 + .8 * (C70 + .8 * (C75 + .8 * (C80 + .8 * (C85 + .8 * (C90 + .8 * (C95 + .8 * (C100 + .8 * (C105 + .8 * (C110 + .8 * (C115 + .8 * (C120 + .8 * (C125 + .8 * (C130 + .8 * (C135 + .8 * (C140 + .8 * (C145 + .8 * (C150))))))))))))))))))))))))))))))) >= .4106248 * (C + 11 / 13 * (C5 + 11 / 13 * (C10 + 11 / 13 * (C15 + 11 / 13 * (C20 + 11 / 13 * (C25 + 11 / 13 * (C30 + 11 / 13 * (C35 + 11 / 13 * (C40 + 11 / 13 * (C45 + 11 / 13 * (C50 + 11 / 13 * (C55 + 11 / 13 * (C60 + 11 / 13 * (C65 + 11 / 13 * (C70 + 11 / 13 * (C75 + 11 / 13 * (C80 + 11 / 13 * (C85 + 11 / 13 * (C90 + 11 / 13 * (C95 + 11 / 13 * (C100 + 11 / 13 * (C105 + 11 / 13 * (C110 + 11 / 13 * (C115 + 11 / 13 * (C120 + 11 / 13 * (C125 + 11 / 13 * (C130 + 11 / 13 * (C135 + 11 / 13 * (C140 + 11 / 13 * (C145 + 11 / 13 * (C150 + 11 / 13 * (C155 + 11 / 13 * (C160 + 11 / 13 * (C165 + 11 / 13 * (C170 + 11 / 13 * (C175 + 11 / 13 * (C180 + 11 / 13 * (C185 + 11 / 13 * (C190 + 11 / 13 * (C195 + 11 / 13 * (C200 + 11 / 13 * (C205))))))))))))))))))))))))))))))))))))))))))

Hi Bruce

Thank you very much for your work.  The Wkyl Macd >0 one seems to work great, thank you.

Looking at the result of the others though, there seems to be some limitations...

For example for Mnthly Macd > Signal Line, the formula using data updated through 24 Apr 14 says

XHB, PG, SCI & DUK Macds are > the signal line and they are below

EWX, GM & WOOD Macds are not > or = to the sigal line, when in fact they are above.

The Wkly Macd > Signal Line formula seems much more reliable, but there are limitations, using data updated through 24 Apr 14 says

KRFT Macd is not > SIgnal line, but in fact it is.

Is there anyway we can amend the formulas get rid of these problems?

Also, how would i code the formula for the Dly Mcd being > the signal line (usual 12,26,9 settings).

Thank you very much in advance, your work is appreciated.

Chris

Hi Bruce,

I am also having great problems with the following which i thought would be simple...one PCF formula for  confirming whether Daily Stochastics is < 30 and one for showing if it's > 70. The settings are 5,3,3 using on exponential average.

Can you please help on this too.

Thank you

Chris

I have tried the various ways think the closest i have gotten to being correct in syntax (but not in result) is the following.

XAVG(STOC5.3,3) <= 30

and

XAVG(STOC5.3,3) >= 70

Thanks in advance,

Chris

I have gone over the formulas for the MACD being above its Signal Line and they appear to be correct.

The intuitive way to write the formula is:

XAVGC12 - XAVGC26 > XAVG(XAVGC12,9) - XAVG(XAVGC26,9)

But it is converted into the following to make the expanded weekly and monthly formulas easier to write while still basically matching the values for a MACD Histogram (which is (( - 136) * XAVGC12 + 24 * XAVGC26 + 112 * XAVGC9) / 51):

(24 * XAVGC26 + 112 * XAVGC9) / 51 > 136 * XAVGC12 / 51

This could actually be simplified to the following, but the above is what was used for the calculations and returns the same results.

3 * XAVGC26 + 14 * XAVGC9 > 17 * XAVGC12

Any difference in what is being plotted on the chart seems to be entirely the result of the need to use slightly different algorithms (the chart is calculated from left to right while the formula is calculated from right to left) and the limited amount of data available for these longer time frames.

There is no way to use the left to right algorithm in a Personal Criteria Formula. The results of the different algorithms converge on the same results with enough data. The long formulas are designed to have enough data to converge on approximately the same value as they would get using an infinite amount of data instead. This ends up being about 90 bars of data of the time frame being averaged. It needs all of this data to calculate at all.

The monthly formula requires at least 1870 bars of data to calculate while the weekly version requires at least 446 bars of data to calculate. So if there are fewer bars than that the formula will not calculate. This means it won't return true (so you might think it is false when it just doesn't have enough data).

The right to left algorithm used by the program can start returning data much earlier, but because it is actually taking a moving average of a moving average, it does not actual converge on the value which would occur with an infinite amount of data until much later.

The raw MACD can start returning data at 26 bars (it is initialized with a simple moving average) and the signal line can start returning data at 34 bars (26 + 9 - 1). That said, it does not converge with similar precision as the right to left algorithm using 90 bars until it has about 120 bars data.

So there are two issues. You have items which appear to be false using the formulas because there isn't enough data to calculate the formula even though there is a plot on the chart. You also have items which might return true in the formula which appear to be false on the chart (or vice versa) because there isn't enough data for the right to left algorithm to converge onto the same result.

The only way to have the formulas start returning data at the same time as the chart starts plotting would be to limit the amount of data used in the formulas. But while these results would match at the left edge of the plot, they would nomally normally match at the right edge of the plot. There just wouldn't be enough data being used in the formulas to correctly converge on the correct result.

It would theoretically be possible to increase the amount of data used in the formulas so that they could not be calculated until there was enough data for the chart to converge as well, but doing so would make the first problem worse as you would now need 2494 bars of daily data for monthly calculations and 595 bars of daily data for weekly calculations.

XAVG(STOC5.3,3) is the 3-period exponential moving average of a 3-period simple moving average of the raw 5-period stochastic.

Understanding Stochastics

If you are interested in the SK of the Stochastic, you just want the 3-period exponential moving average of the raw 5-period stochastic.

XAVG(STOC5,3)

If you are interested in the SD of the Stochastic, what you want is the 3-period exponential moving average of a 3-period exponential moving average of the raw 5-period stochastic.

.50048876 * (XAVG(STOC5,3) + .5 * (XAVG(STOC5.1.1,3) + .5 * (XAVG(STOC5.1.2,3) + .5 * (XAVG(STOC5.1.3,3) + .5 * (XAVG(STOC5.1.4,3) + .5 * (XAVG(STOC5.1.5,3) + .5 * (XAVG(STOC5.1.6,3) + .5 * (XAVG(STOC5.1.7,3) + .5 * (XAVG(STOC5.1.8,3) + .5 * (XAVG(STOC5.1.9,3)))))))))))

So you would just add the <= 30 or >= 70 onto the end of one of the above.

Hi Bruce

Thank you very much for your really helpful and detailed replies. I now have the best formulas i can for what i want. The Daily Mcad > or < Signal line works a treat, as does the Daily Stochastic > 70. 

However, for some strange reason the exact same formula, but with the end changed to <=30 doesn't pick up lots of tickers where clearly the Daily Sto is < 30.

I have ensured i have copied the formula correctly and made the appropriate <=30 amendment and i have no idea why this would perform so differently than its >70 counterpart. Can you please help.

Here are some examples of tickers this formula doesn't pick up for <30. Using data updated through 28 Apr 14 -  KMI, AWR, C, GM, PHM, T.

Thank you in advance,

Chris

I suspect you are interested in the SK of the stochastic instead of the SD of the stochastic based on your example symbols. I have altered the original stochastics reply to add a formula for the SK in addition to the SD. I have also removed one of the SD formulas as I determined that the the remaining formula was definitely the better of the two.

You're correct Bruce and an absolute star! Thank you for yoru help with these formulas and issues. Yes i thought that the remaining stochastic formula performed better too, the updated version you have posted works a charm. Very much appreciated.

Thank you

Chris

You're welcome.