Developing Potential sales Projecting Model with RapidMiner

Rapidminer 5.0 Video Tutorial #8 - Financial Time Series Data Exploration

In this video I show the viewer how to use Rapid Miner’s Time Series plugin to explore time series data. This is a prep for videos #9 and #10 that will teach the viewers how to make financial time series predictions.
Video Rating: / 5

7 Antworten auf „Developing Potential sales Projecting Model with RapidMiner“

  1. Hi !
    It's very interesting and exactly what i need! Could you provide an rmp, please? I have difficulties, reproducing your pipeline.
    Highly appreciated
    Thanks in advance, Andrey

  2. Using the same data, same technique as in the video and RapidMiner 7.4, but the FIT Trend operator gives an error: Missing Predicted Label.
    Please help with any turnarounds.

  3. I get an error…"Missing predicted Label" from the Fit Trend – node. I use db-data and excel data and did exactly the same as in the video (Rapid Miner 7.4).
    Any ideas?

  4. Hello,

    I just started to watch your videos, and it seams to me that they are very well done. 

    I'm new in this field and I'm triend to find out more about how to choose the est algorithm (SVM, NeuralNest..) in some cases. As far as I saw, you are giving some tips in some of your videos..do you have one video that is offering an overview of all these algorithms as well as the best field in witch to use it?

    How should we work with data that is structured but is not numerical (text – like product id, product color…)?

    I would be grateful if you could give me some references,

    Thank you a lot,

  5. Hello Tom, Are you familiar with feature extraction operators in RapidMiner? I will like to use it, but I am stuck with all my attempts. Is there any documentation for Series extension? Thanks.

  6. I think you can understand that if I could use your program to produce things like this 10 to 50 per hour rather than spending 2 to 4 hours PER chart to ensure I got the math right manually I'd be seeing a significant advantage, plus I can then seriously correlate a lot of tickers and see some more patterns of note. At this time I can't produce the math fast enough to do it

  7. I have a question: can the program do something that automates tight fits of moving with the smaller up & down trends? If it can I think I'd like it. So far I've been doing mine manually like for gold or for the S&P. I often use linear regression but I also often add another function like a sine-wave to follow it (or for S&P I'm using a hyperbolic cosine) and on top of that it's all starting with rate of change graphs (doesn't have to but I find them easy to use). e.g. /watch?v=b1mQZAIxC3A

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert.