Shape to the data that is showing

  • Context: Undergrad 
  • Thread starter Thread starter msticky
  • Start date Start date
  • Tags Tags
    Data Shape
Click For Summary

Discussion Overview

The discussion revolves around analyzing a dataset presented in a Google spreadsheet, focusing on identifying patterns or shapes within the data over time. Participants explore methods to determine the load of each column and predict future values based on observed trends.

Discussion Character

  • Exploratory, Technical explanation, Debate/contested

Main Points Raised

  • One participant observes a consistent shape in the data with a leading edge on the left and a lagging edge on the right, suggesting that most activity occurs in the middle.
  • Another participant proposes that finding a function to describe the data is possible, but emphasizes the need for a model to avoid guesswork.
  • A different participant questions the meaning of "model" in this context, seeking clarification.
  • One participant provides an example of a numerical series to illustrate that data points alone may not be sufficient to determine the correct underlying pattern, highlighting the importance of context and expectations for the data.

Areas of Agreement / Disagreement

Participants express differing views on the necessity of a model for analyzing the data, with some emphasizing its importance while others seek clarification on the concept. The discussion remains unresolved regarding the best approach to analyze the dataset.

Contextual Notes

Participants note limitations in the data analysis due to the lack of context about the data's origin, expectations, and potential bounds.

msticky
Messages
12
Reaction score
0
I have a google spreadsheet with a display of numbers from a report that is ordered and the link below is the result of this data.

It seems to me that there is a shape to the data that is showing. This shape seams to be constant over time having the same shape with a leading edge on the left and a lagging edge on the right.

I believe that most of the activity is in the middle of the shape. How can I determine the load of each column that shows which columns are likely to result next?



https://docs.google.com/spreadsheet/ccc?key=0Ajurt2allTaddElHTGZyalZ0aDhpbDBSd1B6UWNoWmc&usp=sharing
 
Last edited by a moderator:
Physics news on Phys.org
I'm sure it is possible to find some function which somehow gives those lines of data (each column is one measurement?), but without any model for the data this is just guesswork and won't tell you much.
 
Not sure what you mean without any Model
 
Last edited:
Well, let's look at an example:
3, 5, 7, how does the series continue?

9, 11, 13, 15, 17... would be an option. But if my series is "all odd prime numbers", this is wrong, and the series continues with 11, 13, 17 without the 9 and 15.

Data points are not sufficient to find (and verify) the correct pattern behind them. It really helps if you know where the numbers come from - what do you expect from those numbers, are there upper and lower bounds to them and so on.
 
Thanks anyway
Link closed
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 169 ·
6
Replies
169
Views
10K
Replies
54
Views
10K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 8 ·
Replies
8
Views
2K
Replies
2
Views
3K
  • · Replies 14 ·
Replies
14
Views
6K
Replies
2
Views
1K
  • · Replies 2 ·
Replies
2
Views
590
Replies
9
Views
4K