1. An open-end fund has a net asset value of $10.70 per share.

It is sold with a
front-end
load of 6%. What is the offering price?
The offering price includes a 6% front-end load, or sales commission, meaning
that every dollar paid results in only $0.94 going toward purchase of shares.
Therefore:
NAV $ 10 . 70
=
Offering price = 1−load 1−0 . 06 = $11.38

2. If the offering price of an open-end fund is $12.30 per share and the fund is
sold with
a front-end load of 5%, what is its net asset value?
NAV = offering price  (1 – load) = $12.30 .95 = $11.69

3. The Closed Fund is a closed-end investment company with a portfolio

currently worth
$200 million. It has liabilities of $3 million and 5 million shares outstanding.
a. What is the NAV of the fund?
b. If the fund sells for $36 per share, what is the percentage premium or discount that
will appear in the listings in the financial pages?

$ 200 ,000, 000−$ 3 ,000, 000

NAV = =$ 39 . 40
a. 5 , 000, 000

Pr ice−NAV $ 36−$ 39 . 40
NAV $ 39. 40
b. Premium (or discount) = = = –0.086 = -8.6%
The fund sells at an 8.6% discount from NAV
 2. The following table lists 12 monthly stock returns (in percent) of a mutual

fund, the market portfolio, and Treasury bill.

Treasury bill Market Fund Month

(r ft) (r mt ) (r pt ) (t)

0.10 2.92 10.38 Jan 2011

0.12 3.82 0.33 Feb 2011

0.10 0.34 0.74 Mar 2011

0.11 2.87 4.61 Apr 2011

0.12 –1.49 –0.53 May 2011

0.09 –1.84 1.55 Jun 2011

0.10 2.25 6.00 Jul 2011

0.12 –5.75 –5.05 Aug 2011

0.11 –2.49 1.72 Sep 2011

0.09 5.40 5.58 Oct 2011

0.08 1.62 –2.23 Nov 2011

0.10 –0.37 –2.19 Dec 2011

Based on the information given in the table, answer the following questions:

(a) Does the fund manager have selection ability?

(b) Does the fund manager have market-timing ability?

Table 1:
Dependent Variable: RP-RF
Method: Least Squares
Date: 12/26/17 Time: 01:23
Sample: 2011M01 2011M12
Included observations: 12

Prob. t-Statistic Std. Error Coefficient Variable

0.2595 1.195453 1.000382 1.195910 C

0.0226 2.692534 0.327069 0.880643 RM-RF

1.639167 Mean dependent var 0.420281 R-squared

4.280453 S.D. dependent var 0.362309 Adjusted R-squared
5.447105 Akaike info criterion 3.418179 S.E. of regression
5.527923 Schwarz criterion 116.8395 Sum squared resid
5.417183 Hannan-Quinn criter. -30.68263 Log likelihood
1.751077 Durbin-Watson stat 7.249738 F-statistic
0.022601 Prob(F-statistic)

Table 2:
Dependent Variable: RP-RF
Method: Least Squares
Date: 12/26/17 Time: 01:24
Sample: 2011M01 2011M12
Included observations: 12

Prob. t-Statistic Std. Error Coefficient Variable

0.3530 0.979352 1.409388 1.380288 C

0.0319 2.536073 0.345117 0.875242 RM-RF
0.8485 -0.196640 0.098749 -0.019418 (RM-RF)^2

1.639167 Mean dependent var 0.422761 R-squared

4.280453 S.D. dependent var 0.294486 Adjusted R-squared
5.609484 Akaike info criterion 3.595362 S.E. of regression
5.730711 Schwarz criterion 116.3397 Sum squared resid
5.564602 Hannan-Quinn criter. -30.65691 Log likelihood
1.681208 Durbin-Watson stat 3.295732 F-statistic
0.084353 Prob(F-statistic)

Table 3:
Dependent Variable: RP-RF
Method: Least Squares
Date: 12/26/17 Time: 01:35
Sample: 2011M01 2011M12
Included observations: 12

Prob. t-Statistic Std. Error Coefficient Variable

0.6442 0.477800 1.945024 0.929333 C

0.2030 1.373082 0.715877 0.982958 RM-RF
0.8741 -0.163007 1.268694 -0.206806 D01*(RM-RF)

1.639167 Mean dependent var 0.421988 R-squared

4.280453 S.D. dependent var 0.293540 Adjusted R-squared
5.610824 Akaike info criterion 3.597770 S.E. of regression
5.732050 Schwarz criterion 116.4956 Sum squared resid
5.565941 Hannan-Quinn criter. -30.66494 Log likelihood
1.826871 Durbin-Watson stat 3.285300 F-statistic
 0.084863 Prob(F-statistic)

(a) A fund manager’s selection ability is judged by the intercept term of the following

(excess return version) CAPM model:

r pt −r ft =α p+ β p ( r mt −r ft ) +ε pt

By estimating the preceding index model, we obtain the estimates of the slope and

intercept:

^β =0.881
p

α^ p=( r p−r f )− β^ p ( r m −r f )

¿ ( 1.74−0.10 ) −0.881× ( 0.61−0.10 )=1.19 %

Note that this estimation can be done by using Excel spreadsheet or any statistical

software package.

Because the Jensen alpha, α^ p, is positive but statistically insignificant, we say that

the fund manager does NOT have superior selection ability.

>> There are other important measures of performance that can also be extracted

from Table 1. These statistics can be used to infer whether the fund is actively or

passively managed. The tracking error measures the performance of the fund

compared to the underlying index. The R2 can be used loosely to indicate how

closely the fund mimics the performance of its underlying index. In our case
2
R =0.42 which is very low indicating that the fund is actively managed. The

Tracking error can be accurately measured by the standard deviation of the

residuals from the CAPM. The Tracking error can be extracted from Table 1 as it
 is the S.E. of regression which is equal to 3.41. also the information ration and the

Traynor measures can be calculated.

(b) According to Treynor and Mazuy (1966), market-timing ability is judged

2
based on the coefficient on the squared term ( r mt−r ft ) , γ p:

2
r pt −r ft =α p+ β p ( r mt −r ft ) +γ p ( r mt−r ft ) + ε pt

By estimating the preceding regression model, we obtain the estimates of

the coefficients:

α^ p=1.38 % , β^ p=0.876 ,∧ γ^ p =−0.02

Here, the magnitude of the estimate γ^ p is very small and it is statistically

insignificant. Thus, we may say that the fund manager does NOT have

market timing ability.

The regression presented in Table 3 is the Henriksson and Merton (1981)

measure in page 673. In short the coefficient on the Dummy variable is

insignificant, we cannot reject the null that the fund manager do not have

market timing ability.

2 Consider the following data for a one-factor economy. All

portfolios are well-diversified.

Portfolio Factor loading Expected

on λ 1 returns
 A 1.2 12%
B 0 6%

Suppose that another portfolio, Portfolio E, is well diversified with a

factor loading of 0.6 and expected return of 8%. Would an arbitrage
opportunity exist? If so, what would be the arbitrage strategy?
The expected return for Portfolio B equals the risk-free rate since its beta
equals 0.
For Portfolio A, the ratio of risk premium to the Factor loading (you
can think of it as a beta) is: (12 – 6)/1.2 = 5
For Portfolio E, the ratio is lower at: (8 – 6)/0.6 = 3.33
This implies that an arbitrage opportunity exists. For instance, you
can create a Portfolio G with beta equal to 0.6 (the same as E’s) by
combining Portfolio A and Portfolio B in equal weights. The
expected return and beta for Portfolio G are then:
2
RG = ∑ X i Ri
i=1

RG =(0.5× 12 %)+(0.5 × 6 %)=9 %

2
b G= ∑ X i b i
i=1

b G=( 0.5 ×1.2 )+ ( 0.5× 0 )=0.6

Comparing Portfolio G to Portfolio E, G has the same beta and higher

return. Therefore, an arbitrage opportunity exists by buying Portfolio G
and selling an equal amount of Portfolio E. The profit for this arbitrage
will be:
R E=8=6+ 0.6 λ1
RG =9=6+0.6 λ 1
RG −R E=1 %

That is, 1% of the funds (long or short) in each portfolio.

Initial CF End of period CF Exposure to
b i1
Portfolio E Short +100 Buy back -0.6
 (1+0.08)100=-108
Portfolio G long -100 Sell 0.6
(1+0.09)100=109
0 $1 0