Forecasting returns and trading strategies

Print Friendly

Mouth hanging open and a look of daydreaming, what’s so amazing to get such a reaction? It’s the prospect of a reliable forecast on a stock’s return! What else? It’s bewildering to still catch an investor’s interested look with the pretence of a cue on a hot stock ready to fly if you consider that manking has undergone tens of worldwide stock bubbles which should have taught a lot.

Nevertheless, if you’re inclined to wishful thinking then you can find in the following stock list the output of a fully automated forecasting algorithm that fits a statistical model to a time series stock returns [1]. As easy as can be by feeding the forecasting method with a time series you get a forecast whose prediction accuracy can disappoint even the light-hearted retail investors flocking the annual ITForum 2014 at Rimini, Italy, a trade fair about financial market trading.

No matter how accurate your forecast can be, you can always rely on your favorite Black-Litterman implementation to smooth out prediction errors and blend your expected returns with the rigor of a mean-variance optimization before logging in on your trading platform and punch in orders to rebalance your portfolio. Before you look back and second guess your prediction skills beware of easy-to-implement-strategies-to-get-rich-without-the-risks-of-getting-poorer.

Ultimately, if a trader whisperer beckons you psst from around the corner and recommends  buying or selling that stock according to that secret unbeaten strategy, it’s not so different from following your old punk friend’s advice to go and get that one-hundred-dollar bill from the dean’s drawer in his unattended office. Think twice.

# Ticker Weight Cum.Weight Volatil. MICER ARIMA BL Sector
1 AAPL 4.65 4.65 1.10 0.01 0.08 0.04 Information Technology
2 XOM 4.00 8.65 1.04 0.01 0.00 0.01 Energy
3 GOOG 3.22 11.87 1.16 0.02 0.16 0.09 Information Technology
4 MSFT 2.96 14.83 1.43 0.02 0.14 0.08 Information Technology
5 JNJ 2.63 17.46 0.91 0.01 0.10 0.06 Health Care
6 GE 2.49 19.95 1.06 0.02 0.06 0.04 Industrials
7 WFC 2.28 22.23 1.04 0.02 0.04 0.03 Financials
8 JPM 2.19 24.42 1.27 0.02 0.11 0.06 Financials
9 BRK-B 2.19 26.61 0.98 0.01 0.09 0.05 Financials
10 CVX 2.16 28.77 0.88 0.01 -0.00 0.01 Energy
11 PG 2.05 30.82 0.81 0.01 0.10 0.05 Consumer Staples
12 PFE 2.02 32.84 1.13 0.02 0.06 0.04 Health Care
13 VZ 1.88 34.72 1.11 0.01 0.00 0.01 Telecommunication Services
14 IBM 1.85 36.57 0.99 0.02 -0.00 0.01 Information Technology
15 BAC 1.82 38.39 1.53 0.02 0.14 0.08 Financials
16 T 1.75 40.14 1.16 0.01 -0.01 0.00 Telecommunication Services
17 MRK 1.55 41.69 0.92 0.01 0.15 0.08 Health Care
18 C 1.47 43.16 1.44 0.02 0.08 0.05 Financials
19 KO 1.42 44.58 0.82 0.01 -0.00 0.00 Consumer Staples
20 AMZN 1.30 45.88 1.67 0.02 0.17 0.09 Consumer Discretionary
21 QCOM 1.28 47.16 1.27 0.02 0.07 0.05 Information Technology
22 ORCL 1.28 48.44 1.29 0.02 -0.04 -0.01 Information Technology
23 DIS 1.27 49.71 1.33 0.02 0.17 0.10 Consumer Discretionary
24 CMCSA 1.26 50.97 1.17 0.01 0.13 0.07 Consumer Discretionary
25 PM 1.26 52.23 0.96 0.01 0.00 0.01 Consumer Staples
26 PEP 1.22 53.45 0.92 0.01 0.00 0.01 Consumer Staples
27 INTC 1.21 54.66 0.87 0.01 0.06 0.04 Information Technology
28 WMT 1.19 55.85 0.83 0.01 0.02 0.02 Consumer Staples
29 SLB 1.18 57.03 1.08 0.02 0.06 0.04 Energy
30 CSCO 1.12 58.15 0.91 0.01 0.00 0.01 Information Technology
31 V 1.10 59.25 1.17 0.02 0.15 0.08 Information Technology
32 HD 1.10 60.35 0.96 0.01 0.08 0.05 Consumer Discretionary
33 FB 1.07 61.42 2.34 0.02 0.45 0.24 Information Technology
34 GILD 1.07 62.49 2.18 0.02 0.19 0.11 Health Care
35 UTX 0.94 63.43 0.98 0.01 0.11 0.06 Industrials
36 MCD 0.92 64.35 1.01 0.01 0.01 0.01 Consumer Discretionary
37 AMGN 0.90 65.25 1.65 0.02 0.22 0.12 Health Care
38 CVS 0.87 66.12 1.00 0.01 0.15 0.08 Consumer Staples
39 UNP 0.84 66.96 1.03 0.02 0.17 0.09 Industrials
40 BMY 0.83 67.79 1.73 0.02 0.11 0.07 Health Care

 

BL-MIC-1002014-03-23

[1] Automatic Time Series Forecasting: The forecast Package for R by Rob J. Hyndman and Yeasmin Khandakar both of Monash University, Journal of Statistical Software, July 2008, Volume 27, Issue 3.

 

Share Button
Posted in Economics
Tags: , ,

Leave a Reply

Your email address will not be published. Required fields are marked *

*

47 + = 51

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Tweets