GDX

[Pascal]

GDX gave back already what it provided yesterday, leaving us with a general positive divergence in a "strong buy" environment.

In general, the long-term picture (3 months) shows that we have been in a consolidation mode since last November. There is a good support below us and the stop level gives us much room. There is also strong resistance above with the 50/200MA combination.

I will be adding to the long position when I see a positive intraday divergence being formed (MF dropping less than the price and showing sellers' temporary exhaustion)



Pascal

17030

17029

[Pascal]

The negative MF accelerated late in the day yesterday.
If we cross below the dotted cash level, we will turn to cash.
The model will not short because the ATR level is very low.

As we can see below, in the past, bounces in ATR have been experienced when investors were afraid of a pull-back.
I do not believe that investors are afraid of a pull-back. However, the sector is not as "hot" as most other sectors and the type of reversal that we had yesterday is not good at all.



Pascal

17047

17048

[Harry]

Hi Pascal,

I have been experimenting with different variations of ATR and wanted to inquire what formula you prefer? Do you utilize the Stockcharts formula:

http://stockcharts.com/help/doku.php?id=chart_school:technical_indicators:aver age_true_range_a

or another variation for your calculations? I have tried a few but assume you filtered down to the most effective.

Thank you,
Harry

[Pascal]

Hi Pascal,

I have been experimenting with different variations of ATR and wanted to inquire what formula you prefer? Do you utilize the Stockcharts formula:

http://stockcharts.com/help/doku.php?id=chart_school:technical_indicators:aver age_true_range_a

or another variation for your calculations? I have tried a few but assume you filtered down to the most effective.

Thank you,
Harry


We have been using a 20D ATR, which is the average absolute of the price difference between the Max of (High of day, Previous close) and the Min of (Low of day, Previous close).

We did not "optimize" but measured different ATR periods and the 20D seemed to be the most adequate, although it is slow.


Pascal