- UID
- 223233
- 帖子
- 477
- 主题
- 143
- 注册时间
- 2011-7-11
- 最后登录
- 2013-9-26
|
4#
发表于 2012-3-23 16:10
| 只看该作者
Xavier Fellows works in the research department of Multinational Inc., a large investment bank. He is tasked with forecasting economic conditions to support the bank's money managers and traders.
Fellows takes his work seriously and is considered to be an excellent forecaster. His economic forecasts are updated monthly and sent to most of Multinational’s analysts and money managers. The analysts use Fellows' forecasts as the basis for their own research on specific securities or asset classes.
However, Fellows is concerned that his forecasts are not accurate enough. In an effort to avoid making mistakes, Fellows follows a detailed process to develop accurate and usable forecasts. Fellows hopes that this process will help him avoid some of the common problems of forecasts. Here is his system:- Establish a benchmark for market expectations. Multinational serves thousands of clients with different investment goals and constraints, and Fellows knows analysts will need the different benchmarks for a variety of different types of investors.
- Look at the historical returns of a number of asset classes to act as a check on forecasts for each asset class.
- Assemble data on historical returns and valuations for all relevant asset classes, considering potential biases, adjusting the numbers to account for different calculation methods, and ensure that data definitions match those used by the company that collected the data.
- Interpret the data. Fellows uses his years of experience to extrapolate that data into growth and valuation assumptions for each asset class. This step is the most subjective.
- Distill assumptions into top-down forecasts, detailing the assumptions and methods for interpreting historical data in the event that individual analysts want to use data to create their own industry-specific forecasts.
- Monitor performance. If Fellows’ forecasts prove to be inaccurate, he works to improve his models.
This month's forecast dwells heavily on inflation projections and their expected effect on the returns of different asset classes. Fellows projects a decline in inflation and predicts that bond yields have bottomed out.
Stock analyst Karen Andrews calls Fellows after the report is released with some questions about his analysis. She is pleased with the work, but a bit disappointed that he did not include information on current and estimated bond yields.
Andrews asks Fellows to forward his analysis of the inflation picture to Carol Huggins, a colleague who works in the bank's money-management business. Huggins consults on money-management issues with large clients and is very interested in inflation projections.
Lester Canfield, who manages money on a discretionary basis for high-net-worth individual investors, is also interested in Fellows' forecast. After reading the entire document, he decides to sell some of his clients' interest in a limited partnership that develops and manages real estate, and invest that money in high-yield bonds. Canfield's reasoning is threefold: - Canfield believes the partnership has excellent return potential, but he is the only one who expects such robust results. The bonds seem to be a safer investment, and Canfield does not want to guess wrong.
- Historically, average high-yield bond returns are higher than the returns of real estate partnerships.
- During periods of falling inflation, real estate investments often lag the market.
Before making the move, Canfield calls Fellows to get an opinion on his plan. After hearing Canfield's rationale, Fellows advises against the move into high yield bonds.Fellows skipped a step in his technique for producing forecasts. He forgot to: A)
| identify a valuation model used in his analysis. |
| B)
| identify where he obtained his data. |
| C)
| assure that the underlying data is accurate. |
|
Fellows' plan mirrors the seven-step process for formulating capital-market expectations in every aspect except one, identifying the valuation model used in the analysis. Assuring the accuracy of data and identifying its source are important, but they would presumably fall under steps three and five of Fellows' process. (Study Session 6, LOS 18.a)
Fellows' advice to Canfield suggests Canfield is least likely suffering from: | B)
| the recallability trap. |
| C)
| failing to use conditioning information. |
|
The relationship between historical returns and economic variables is not constant over time, and Canfield may not be considering information about changing economic conditions that affected real-estate returns over short periods of time. Analysts fall into the prudence trap when they become overly conservative because they are afraid of being wrong. The use of ex post (after the fact) data to interpret ex ante (before the fact) actions is risky. There may be other factors, whether correlated with inflation or independent, that caused subpar real estate returns. The recallability trap has to do with allowing dramatic events to affect forecasts. This issue is not relevant here. (Study Session 6, LOS 18.b)
Andrews most likely requested bond yields because she wanted to: A)
| analyze stock-market valuations using the risk premium approach. |
| B)
| develop a shrinkage estimate. |
| C)
| gauge potential fixed-income investments. |
|
The risk premium approach uses bond yields and an equity risk premium to project market returns. Since Andrews is an equity trader, it is unlikely she is interested in fixed-income investments. The question of shrinkage estimators is not relevant here. (Study Session 6, LOS 18.c)
Which of the following is least likely a common problem encountered in forecasting? A)
| Data measurement errors and biases. |
| B)
| It is difficult to use multiple regression analysis. |
| C)
| Failing to account for conditioning information. |
|
There are nine problems in producing forecasts: - limitations to using economic data
- data measurement error and bias
- limitations of historical estimates
- the use of ex post risk and return measures
- non-repeating data patterns
- failing to account for conditioning information
- misinterpretations of correlations
- psychological traps
- model and input uncertainty
Due to the problem of misinterpretation of correlations, it is often useful to run multiple regressions. An analyst may discover a stronger relationship between two variables that was not evident using simple linear regression analysis. (Study Session 6, LOS 18.b)
Due to the decline in inflation and the low bond yields, Fellows should conclude that the economy is most likely in what stage of the business cycle?
In general, inflation rises in the latter stages of an expansion and falls during a recession and the initial recovery. Bond yields peak during a slowdown and fall during a recession, however, they bottom out during the initial recovery stage. (Study Session 6, LOS 18.e)
Which of the following is least accurate regarding inflation? A)
| Declining inflation results in declining economic growth and asset prices. |
| B)
| Highly levered firms are most affected by declining inflation rates. |
| C)
| Low inflation affects the return on cash instruments. |
|
Low inflation can be beneficial for equities if there are prospects for economic growth free of central bank interference. Declining inflation usually results in declining economic growth and asset prices. The firms most affected are those that are highly levered because they are most sensitive to changing interest rates. Low inflation does NOT affect the return on cash instruments. (Study Session 6, LOS 18.g) |
|