THOMPKINS FINANCIAL ADVISORY SERVICE |
WHAT TO BUY (FUND SELECTION) |
The buy function requires the conducting of significant technical analysis in order to determine the Fidelity sector fund that has the greatest probability of being profitable and that also will be amongst the best performing funds at the end of the buy/sell cycle. This process employs multiple filters which cause many of the funds to be eliminated so that only the funds with the greatest profit potential remain in the universe of funds under consideration. The activities described below are the major steps that must be completed at the end of the day after detecting a buy signal in order to arrive at the Fidelity sector Fund that should be purchased. |
Assign each one of the Fidelity sector funds to one of seven sector fund family group. The seven family groups are Technology, Finance, Health, Natural Resources, Consumer, Cyclical Industries and Utilities. Rank the families in accordance with each family performance. Filter -1 (Identifies and removes any fund that is not in an upward trend} Compute for each fund the five moving averages defined earlier and develop a moving average combination function that shows where each moving average is (greater than closing price, less than closing price or equal to closing price) in relationship to each fund closing price. This methodology is proprietary and details are not provided. Identify the valid moving average combination functions, one of which a fund must meet, in order for a fund to be considered for purchase. (Each valid moving average combination function must indicate a rising trend). Remove any fund for which its moving average combination function is not equal to one of the valid moving average combination functions. Filter -2 (Identifies and removes any fund that is oversold or overbought} Remove any fund that is Oversold or Overbought. A fund is Oversold if its recovery level, following a market downturn, lags the recovery level of its peers as well as the indices that caused the buy signal to be generated. A fund is Overbought if its recovery level, following a market downturn, exceeds the recovery level of its peers as well as the indices that caused the buy signal to be generated to the extent that there is a high probability the fund will begin to experience an immediate down turn. The methodology to determine when a fund is Oversold or Overbought is proprietary and is not provided. Filter -3 (Identifies and removes any fund that lags the performance of its peers) Identify, using analytical techniques, the short term market cycle, the intermediate term market cycle and the long term market cycle. Determine the average number of days that makeups each of three market cycles . Compute, for each fund, its performance for the long term and the intermediate term market cycles. Identify the fund(s) that had the best performance for the long term and the intermediate term market cycles. Compute 50% of the percentage value identified for the fund with the best performance for the long term market cycle and 50% of the percentage value for the fund that has the best performance for intermediate term market cycle. Remove any fund from further consideration where its long term market cycle performance and its intermediate term market market cycle performance are less than the associated percentage values computed in the previous step. Filter -4 (Identifies and removes the entire family group if there are not two or more funds remaining in the same family group after completing the above analysis) Remove any family group and associated fund that has only one remaining fund. Our analysis led us to the conclusion that a fund, purchased from a family group that has multiple funds remaining after completion of the filtering process as described above, has a higher probability of being profitable during a buy/sell cycle than a family which has only one fund present. Compute, for each remaining fund, the rate of change (ROC) measured from the day the fund reached its lowest value during the current short term market cycle to the day of the buy signal. Compute a raw score (Ranking) for each remaining fund using its performance value derived for the long term market cycle, the performance value derived for the intermediate market cycle and the ROC value. The methodology employed to implement this task is proprietary and details are not provided. Select the fund for purchase using the Ranking developed for each remaining fund and the priority of the fund families |
">
">
">
">
