Options

Blog - The stochastic resonance program (part 2)

This is not yet ready to go, but quite close -- the whole intent and flow of the article is spelled out here.

I still need to work on the intro, the conclusion, and to actually check the coding steps that I describe.

Comments

  • Options

    This article is ready for content review.

    I'll want to make one more pass over the form before publishing it.

  • Options
    edited August 2014

    Removed "conclusions" which was essentially a plug for more participation in the Azimuth code project. We're in a new chapter now that we have an active project. I still would like to work on outreach towards developers, but I should rethink the message our new context.

  • Options

    The articles looks fine to me now except for this:

    sindot = the sum product of sinwave and the output signal

    Did you mean "sum" or "product"? I don't know what "sum product" means.

    I agree that we're in a somewhat new chapter now that we have an active project that can soak up all our energy - Experiments in El Niño analysis and prediction. I think we should mainly try to get coders to help us with this!

    So, I think it may be good to gracefully wrap up the stochastic resonance project... at least for now. (We can always return to it.)

  • Options

    Anyway, except for that one "sum product" issue I think this one is good to go, and I'd like to post it as soon as possible!

  • Options

    "Sum product" (sum of products) is a neologism for dot product, which hit the big screen when Excel used sumproduct as the name of the function. But context is everything, so I'll stick with the known term dot product.

    So, let do one more editing pass over this, I'll let you know when it's good to go. I promise not to belabor it, but still, I want to give the writing one more look-over.

  • Options

    Okay, I'm desperate for new stuff on the blog!

  • Options

    David - what about this blog article? I'm eager to post it ASAP.

  • Options

    Ok -- good to go.

    I did some minor editing.

    Thanks

  • Options

    Please consider adding a definition of what a sample path instance is and explaining what we get from adjusting it? Cheers

  • Options

    Ok, fair point. I changed the sentence to this:

    Change the Sample-Path slider to alter the sequence of random numbers that are fed to the process. This will cause a different instance of the process to be displayed.

  • Options

    Thanks, I will post it now!

  • Options
    edited August 2014

    Okay, I posted it here:

    Since it's been so long since part 1, I added a little text and a picture near the top to help remind people of the idea. I also made it easier to find the online program. Feel free to demand changes here, David!

    Thanks for writing this! It's great to let people "peek behind the scenes" and see how software actually works.

  • Options

    Great editing job, thanks! The picture and the more peppy wording in the beginning are very helpful.

  • Options

    While there haven't been many comments on this post yet, it hasn't gone unnoticed: the number of views spiked to 200/hour for about 8 hours when someone on Reddit linked to it. Now I've finally linked to it from Google+, which should get more views, perhaps from more talkative people.

  • Options
    edited September 2014

    About the stochastic differential equation app itself, Marshall Hampton wrote:

    Its frustrating that the app doesn't actually show the stochastic differential equation its illustrating - what is it exactly?

    I agree; we shouldn't be reluctant to show the differential equation itself:

    $$ d X(t) = (X(t) - X(t)^3 + A \; \sin(t)) \; d t + \sqrt{2D} \; d W(t) $$ where $A$ and $D$ are constants that can be adjusted on the sliders, and $W(t)$ is white noise. I think we should include the equation somewhere in this app, with a link to Glynn Adgie and Tim van Beek's explanation.

    I guess the app is on Glynn Adgie's website. Should I contact him or just put a modified copy on my own website?

Sign In or Register to comment.