The Joy of Stats

There was a programme on BBC4 the other night called the 'Joy of stats'. presented by a light hearted professor, it gave a run down, very briefly, on the use of statistics and the relevance as well as the way they are used nowadays. Including Google's translator system.

Whilst its undoubtedly not a new thought, the amount of data generated by services like WSPR and DX cluster could feasibly be an alternative method for generating almost live propagation prediction. I'm not a great user of these services as I quite like having a twiddle to see 'whats out there'. I can see a use for this data. At one point I tried to do some data analysis on the 3 years worth of information I have gathered from listening (through Faros) to the IBP beacons. Not having either a clue what to do with it I could do little else but generate a few graphs that presented the data in a slightly different way. Not much use either. The stock presentation is great. but a correlation between what is heard and what the sun is doing and what you are expected to hear from the likes of VOACap is interesting to say the least.

There is clearly scope for using this data to give an alternative view for prediction. The model would no doubt be complex and likely to be as reliable as other mathematical models. My experience with commissioning a and running mechanical based simulation is that if you get 60% accuracy then you're doing well. Still worth trying another angle, not least of which just for the sheer hell of it.

Veering into 'off topic'-ness as this is I still see the draw of modelling and simulation. Even though I know its rarely as accurate as we'd like it to be. I do seem to stray into areas which i don't have the required skills, so excuse the babble and be thankful I don't do this every day :-D