1. Research and contribution

Economist Harry

Markowitz won the Nobel Prize in 1952 for developing the Markowitz Asset

Allocation Model in his paper “Portfolio Selection”. The model is also known as

Modern Portfolio Theory, or can be simply abbreviated as MPT which I will use

for simplification throughout this thesis. Before MPT, there does not exist a

systematic way for fund managers to allocate weights to their assets or asset

classes. Furthermore, MPT greatly emphasizes on the benefits of

diversification, which means the investors can reap the benefits of investing

in multiple assets or asset classes in specific weights.

There are quite a

few researches available about the application of MPT. These researches,

however, only focus on finding the optimal portfolio consisted of individual

stocks, using the Markowitz Model. Therefore, my research topic about whether

or not a fund manager can beat the market benchmark with the Markowitz Model,

using the empirical data from the Scandinavian stock market, digs more in depth

into the subject by comparing the results with the market benchmark and more

specifically using the Scandinavian market. And lastly, as there is an ongoing

debate on whether or not fund managers can actually beat the market in the long

run, the result from this thesis will provide economic significance for both

fund managers and investors. More and more active fund managers are turning to

passive index fund managers as the historical data over the last 15 years show

that most fund managers fail to outperform the market in the long run and even

when they do, it is due to luck. This thesis is therefore to challenge this

statement by exploring whether a fund manager can outperform the market, using

specific data from the Scandinavian market. The results of the thesis will not

only give an answer to this statement but also provides values for fund

managers and investors. From the perspective of fund managers, if the result

shows that it is a fruitless attempt to try to outperform the market, the fund

managers might want to change from active fund management to passive fund

management. On the other hand, from the perspective of investors, the results

will give them a better idea on which type of fund managers they should really

be looking for.

2. Research method

I will use the existing theories

and the underlying assumptions of MPT as the foundation for my approach to my

research question. It is assumed that the future will be like the past (Sharpe,

2000). Therefore, the future values of average return, risks and correlation

can be calculated based on the historical values. However, it should also be

noted that for the above assumption to hold, the requirement that the

underlying distribution of return has to be stable over the time period

chosen(Sharpe,2000). Therefore, I have chosen a 10-year time frame from

1990-2000, where the distribution of return appears to be quite stable. The

10-year period has to be divided into two parts. The first part is the first

five years (1990-1995), which will be used as the estimation period on which we

will base our future prediction, and the second part is the second five years

(1995-2000), which will be used as the investigation period to recalculate

optimizing weights to each sub-index quarterly.

The benchmark index we chose is

the VINX Nordic Benchmark Equity Indexes, produced

in conjunction with Oslo Börs (Oslo Exchange), track constituents from each of

the Nasdaq Nordic exchanges (Copenhagen, Helsinki, Oslo, Reykjavik and

Stockholm) and Oslo Börs(The benchmark index is consisted of 10 sectors which

we will later use as our sub-indexes to construct optimal portfolios to be

compared with the benchmark.

A portfolio, consisting of 10 sub-indexes that are

compounded in a specific way according to MPT, is used to compare with the market

benchmark VINX. The 10 indexes chosen are the following: basic materials,

consumer services, consumer goods, financials, health care, industrials, oil

gas, technology, telecommunication services, utilities. Each of these indexes

represents a sector or industry in the Scandinavian region. The

ten-sector breakdown is based on the VINX industrial classification of 36

industries. Each sector may be regrouped if a change in industrial structure

makes adjustment necessary. The 450 most liquid issues will be categorized into

the 10 sectors mentioned above(Nasdaq,2018). The fact that these indexes are

major components of the VINX index makes the relevance and the analysis from

the research accurate. The historical data for each of the sector and the

benchmark index is provided by NASDAQ Nordic and can be found on its website. I

have chosen to obtain all the data denominated in Euro, and all the data in

this thesis is daily data unless mentioned otherwise.

In order to construct the capital

market line, we will also need a risk free asset. The risk free asset I chose

is a 3-month treasury note from Riksbanken.

A weighting process needs to be

added in the evaluation process. Because of the fact that macro-economic

factors affect the market and a specific sector fluctuates over time (De Bondt,

1985), distribute returns are not fixed over time and therefore the weights of

the portfolio have to be recalculated based on the average returns and

correlations among the 10 sub-indexes. New estimates for the average returns

and standard deviation of the risky portfolio are recalculated every quarter in

the investigation period using the preceding five years of historical data(Konno,1991).

A new optimal risky portfolio is constructed based on the new weights assigned

to each sub-index each quarter. The optimal weights for the portfolios are

presented in the appendix.

The returns from the portfolio

will be compared to the returns of the benchmark VINX. One thing that needs to

be noted when evaluating performances is the fact that two portfolio have

different risks and therefore the risk-adjusted returns have to be taken into account

(Konno, 1991). No valid results can be obtained without such risk

adjustments. Sharpe ratio is a useful

measure in this case since it indicates the reward-to-variability ratio, or in

other words, it examines the performance adjusted for risks. A Sharpe ratio

graph that covers 20 quarters in the investigation period (1995-2000) will be

constructed for both the VINX benchmark and the optimal portfolios. To further

investigate whether the portfolio outperforms the market benchmark VINX, I will

calculate the risk adjusted returns for both VINX and the portfolios. In order

to compare the returns in a solid way, I will match the risks of the portfolio

weights with the risks of the benchmark, and then present them in a

risk-adjusted returns graph and compare the returns. The risk adjusted matrixes

are presented in the appendix.