Can Copper Predict Short term Movements?

Discussion in 'Metal Futures' started by etfarb, Mar 25, 2013.

  1. etfarb

    etfarb

    i've ran a few regressions with copper and have gotten some extremely interesting numbers.

    I want to make a model with copper being the lead. My biggest question though is can copper actually be useful in day to day or week to week trades?
     
  2. bone

    bone ET Sponsor

    Copper used to be highly correlated with alot more commodities and equity stocks than it is at present. If you run a 2 year correlation looking at daily closes, there isn't much above 90 % positive correlation to speak of really. Again, that used to not be the case - but it is reality at present.
     
  3. Maverick74

    Maverick74

    You can thank China for that. I suspect that correlation between copper and FXI is higher.
     
  4. etfarb

    etfarb

    It is. Infact the hang seng moves in relation to copper

    The question is, can the movement of the hangseng help a speculator make a trade on American Indices?
     
  5. Maverick74

    Maverick74

    I suspect not. Very DIFFERENT markets. Honestly China and the US have nothing in common. Not to mention China is the most corrupt country on earth.
     
  6. etfarb

    etfarb

    Find it hard to hear you say that considering China buys up all of Americas debt, making there a direct relation

    Regardless

    if you were to make a predictive model of the S&P500, what variables would you include/test?
     
  7. Maverick74

    Maverick74

    China does not buy up all our debt. They own 10% of it. About 1.1 trillion. You know who else does? Japan! They both own about 1.1 trillion. And I don't think Japan has been very correlated to our markets in over two decades.

    China is a commodity country. The US is not. They just aren't comparable. In fact, China has been one of the biggest manipulators in the copper market being they own about 40% of the world supply of Copper. So if China slows down, and they have slowed down, copper prices will be weak and they have been weak.

    There are way too many variables that can be predictive models for the s&p. But to play along I would say two of the better ones are the Libor-OIS spread and the yield curve both 2/10 and 10/30.