Skip to content

A Simple Estimation of Bid-Ask Spreads from Daily Close, High, and Low Prices

10 messages · Henrique Ramos, Adam Ginensky, Brian G. Peterson +4 more

Original
diego peroni
# 
Hi All,

I?m looking for a function in R to estimate Bid/Ask Spreads of stocks using Daily candlesticks.

Can anyone suggest some implemetations?

Thanks
Diego
Brian G. Peterson
#
On Tue, 2020-07-28 at 19:40 +0200, diego peroni wrote:
Diego,

I would like to help you, but what you are asking for is simply impossible.

Daily Range, Volume, and Volatility tells you nothing about the intraday numbers except their upper/lower bounds.

It certainly doesn't tell you anything about intraday spreads.

You can certainly *guess* that less liquid instruments have larger effective spreads, since this is usually the case, but you can't know *what* the spread is from daily data.

Brian
Henrique Ramos
# 
Hi there,

You can calculate the Corwin-Schultz (CS) spread estimator.
https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1540-6261.2012.01729.x

One of the authors provides spreadsheets with calculations in Excel. It
should not take much effort to put in R.

You should notice that low-frequency proxies are highly subject to
estimation errors.

Em ter., 28 de jul. de 2020 ?s 14:40, diego peroni <
diegoperoni1971 at gmail.com> escreveu:

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Adam Ginensky
    

    

      
      #
    

  


  

      
To expand on what Brian said, imagine a very illiquid stock that just
trades once on the bid.  No range, but probably a very wide bid/ask spread.
On the other hand imagine a very liquid stock highly correlated to the
market on a day with a large range- still will have a small bid/ask spread.

On Tue, Jul 28, 2020 at 1:09 PM Brian G. Peterson <brian at braverock.com>
wrote:

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Ajay Shah
    

    

      
      #
    

  


  

      
perhaps something like:

https://onlinelibrary.wiley.com/doi/full/10.1111/j.1540-6261.1984.tb03897.x

?

is easy to write.


On Tue, 28 Jul 2020 at 23:10, diego peroni <diegoperoni1971 at gmail.com>
wrote:

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Brian G. Peterson
    

    

      
      #
    

  


  

      
Ajay,
The method you are proposing is easy to write, the authors flat out say
that they didn't really bother to check how accurate their measure is,
but what little checking they do gives highly *implausible* results.
 They also say that it only makes sense for liquid instruments (for
which intraday data is readily available anyway).  So I'm not sure this
makes the impossible any more possible.
Brian

        
On Tue, 2020-07-28 at 23:50 +0530, Ajay Shah wrote:

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Jasen Mackie
    

    

      
      #
    

  


  

      
I am not aware of any bid-ask spread estimators in R. I would be interested
to know the intention with estimated bid-ask spreads (and how they compare
with actual spreads) as they do sound dangerous if they were to be used in
any system/model. The quoted papers are quite old, one dating back to 1984.
The need for estimating bid-ask spreads was likely more necessary then than
it is now when you can easily extract and store this information if you
have access to the data.

        
On Tue, 28 Jul 2020 at 14:21, Ajay Shah <ajayshah at mayin.org> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Alec Schmidt
    

    

      
      #
    

  


  

      
Brian,
You're right, of course. But the Roll's model was an influential work in 1980s when the bid/ask prices were not easily available (if at all). But the transactional prices were available ( 'time and sales' tapes). So, this model was a nice and useful theoretical exercise. ?
Alec

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      diego peroni
    

    

      
      #
    

  


  

      
Thanks Brian and others!

I?m convinced that it?s not accurate but it can give me an idea of the magnitude in particular trading many stocks (global mean).

I?ve found just this post:

https://stackoverflow.com/questions/45916124/estimate-bid-ask-spreads-from-daily-high-and-low-prices-corwin-schultz-2012 <https://stackoverflow.com/questions/45916124/estimate-bid-ask-spreads-from-daily-high-and-low-prices-corwin-schultz-2012>

Regards
Diego

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Ajay Shah
    

    

      
      #
    

  


  

      
These methods are also useful in studying the past, where good datasets are
unavailable. As an example, when Corwin/Schultz first came out, I had
written
https://blog.theleapjournal.org/2012/04/new-insights-into-events-on-indian.html


On Wed, 29 Jul 2020 at 13:05, diego peroni <diegoperoni1971 at gmail.com>
wrote: