help. Plot shows blank when cant compute value.. isnan not working for me

[drive9]

I am trying to get an exception going where when the plot cant compute, to show a 0 or a 1 instead of blanking the whole indicator chart.
Here I'm attaching my code. Using ticker ACIU on todays , 08/31/2021 , it shows fine for length 5, 50, but blanks when using 5000.

declare lower;
def price = close;
def length1 = 5000; # 5, 50 work, 5000 blanks it.

def prelrs1 = ( WMA(price, length1) -  Average(price, length1) ) ;
plot LRS1 = if isnan(prelrs1) then 1 else prelrs1 ; # here Im trying to catch the exception to make it plot 1 when error.
plot zeroline = 0;

 

[mashume]

I doubt whether you'll have much luck getting a moving average of that length to work at all. (5000 minutes is 83 1/3 hours or nearly 3 1/2 days -- which is a LOT of 1 minute candles. If you had 5000 days on the chart, you'd be looking at nearly 20 years of data at 253 trading days per year.) You'd be better off, in my opinion at least, doing something like using a secondary aggregation to get a moving average over that length of time.

def agg = AggregationPeriod.HOUR; 
def data = close(period = agg);
input length = 83;
def prelrs = ( WMA(data, length) -  Average(data, length) ) ;

details of the periods available here:
https://tlc.thinkorswim.com/center/reference/thinkScript/Constants/AggregationPeriod
This may, or may not give you what you want at all, but it will likely give you something.


Happy Trading,
Mashume

 

[drive9]

I doubt whether you'll have much luck getting a moving average of that length to work at all. (5000 minutes is 83 1/3 hours or nearly 3 1/2 days -- which is a LOT of 1 minute candles. If you had 5000 days on the chart, you'd be looking at nearly 20 years of data at 253 trading days per year.) You'd be better off, in my opinion at least, doing something like using a secondary aggregation to get a moving average over that length of time.

def agg = AggregationPeriod.HOUR;
def data = close(period = agg);
input length = 83;
def prelrs = ( WMA(data, length) -  Average(data, length) ) ;

details of the periods available here:
https://tlc.thinkorswim.com/center/reference/thinkScript/Constants/AggregationPeriod
This may, or may not give you what you want at all, but it will likely give you something.


Happy Trading,
Mashume

thanks for the input @nashume. Let me clarify, I dont plan on using a 5000 moving average, it was just an example to show that the variable of 5k breaks the plot. I want to totaly void this variable/plot when a null plot is generated, not to draw it in a different aggregation time.
if I have 3 or 10 moving averages going on , and one of them happens to be null, the entire indicator script will show blank along with the other 9 moving averages. I want the other 9 MA to still show even if 1 or two plots have a variable that rendered NA.

 

[mashume]

thanks for the input @nashume. Let me clarify, I dont plan on using a 5000 moving average, it was just an example to show that the variable of 5k breaks the plot. I want to totaly void this variable/plot when a null plot is generated, not to draw it in a different aggregation time.
if I have 3 or 10 moving averages going on , and one of them happens to be null, the entire indicator script will show blank along with the other 9 moving averages. I want the other 9 MA to still show even if 1 or two plots have a variable that rendered NA.

I see.

There isn't a way. By the nature of ThinkScript, when one plot breaks, they all break. ThinkScript is a functional language rather than procedural. What you're trying to do is more likely to work in a procedural language. It's not that the value is NaN, or anything else... it just broke with your value. When that happens, it's kind of like a kernel panic for this one function and it just breaks. ToS doesn't have error trapping in the sense of a pythonic "Try: ... Except: ..." statement.

You can, of course, put multiple plots on a single chart by overlapping them, but that may not do what you need.

-mashume

 

[drive9]

I see.

There isn't a way. By the nature of ThinkScript, when one plot breaks, they all break. ThinkScript is a functional language rather than procedural. What you're trying to do is more likely to work in a procedural language. It's not that the value is NaN, or anything else... it just broke with your value. When that happens, it's kind of like a kernel panic for this one function and it just breaks. ToS doesn't have error trapping in the sense of a pythonic "Try: ... Except: ..." statement.

You can, of course, put multiple plots on a single chart by overlapping them, but that may not do what you need.

-mashume

This is the answer. Thank you so much Mashume. I will section my code so that section fails on its own.