Kudos to MMs
- Thread starter nitro
- Start date
You seem like a bright guy. My model takes in unemployment as a buy signal, since unemployment mean reverts. However, this time it hasn't and people are calling it a new normal. I know you use a neutral net, but do you know then effect of each node in an economically intuitive way?
A separate topic. Do you calculate Sharpe and percentage up days? I have been working on the mathematical relationship between them. What does it mean when they disagree? Non-normality or dataming?
A separate topic. Do you calculate Sharpe and percentage up days? I have been working on the mathematical relationship between them. What does it mean when they disagree? Non-normality or dataming?
With a heavy heart I have to say that high unemployment is likely a new normal. This economy was sustained by so much retail and financial services it was scary. There is such huge overcapacity and redundancy in retail that a whole slew of people may still lose their jobs. Still, shopping is as American as apple pie. Financial services are also being scaled back in droves with people exiting the stock market and the housing market in shambles.
Where there are prospects for employment, high tech jobs and health care related jobs, I see many young Americans going into nursing - a smart move. On the other hand, I see very few Americans going into more science related jobs like programming, semi-conductors, biology, etc. It seems as if these jobs are going to Asians and Indians. The whole notion of work is getting really strange, and the new normal is that you may have to reinvent yourself at least twice in your life.
To me the notion of an eigth hour work day is an anachronism. What should happen is less hours for more people, say five for one person and three for another. That would put more to work, shifting our economy to a part time work economy. Less hours, more getting payed, control inflation, and you have a pretty good society, to say nothing that it is more healthy - e.g., we don't sleep enough, but we would with a couple of more hours a day that is ours.
I use a NN? That is news to me.
No, I don't. There are so many reasons why similar things would diverge on the same data. Some are artifacts of the computation itself, and some although far more rarely, the data is telling you something.