### Table 1: Optimal strategy in percentage with ES , = 1%. We impose short-sale constraints. All other 15 stock has not been selected.

2002

"... In PAGE 21: ... We compute the optimal strategies for di erent values of and a given xed . Table1 shows the results for some level of and Figure 5 gives the e cient frontier with and without risk-free asset. The tangency portfolio from the ( ; )-analysis corresponds to the tangency portfolio in the ( ; ES1%)-e cient frontier, as one would expect in the multivariate Gaussian case.... ..."

Cited by 1

### TABLE 6. Stock returns and trading volume .fl D 0:98, D 1:5/

### Table 1 Sales of computer monitors.

2005

"... In PAGE 1: ... This article considers verbalization of more complex frequency constraints: multi-role frequency constraints (applying to sequences of two or more roles); and external frequency constraints (involving more than one predicate). Verbalization of multi-role frequency constraints Consider the report displayed in Table1 , which lists an extract of details about sales of computer monitors. In this domain, cities may be identified simply by their name.... In PAGE 2: ... For example, a frequency constraint of 2 on the first two roles of the fact type from Year to Year the number of staff of Gender changed by Quantity may be verbalized thus: Each Year1, Year2 combination occurs zero or 2 times in the population of from Year1 to Year2 the number of staff of Gender changed by Quantity . Figure 2 displays a class diagram in the Unified Modeling Language (UML) notation [13] for our Table1 example. The population is omitted because UML provides no convenient way of displaying it.... ..."

Cited by 4

### Table 4: Decomposition of the unconditional book-to-market (U.S. data) This table reports the unconditional variance decomposition of firms apos; log book-to-market ratios (BE/ME) computed from the U.S. data sample (1937-1997, 192,661 firm-years) for three groupings based on either firm size (Panel A) or firm BE/ME (Panel B). We calculate yearly market capitalization breakpoints for the size groupings in Panel A using only NYSE stocks. The first column shows the horizon N. Within each grouping, the first column shows the simple predictive regression coefficient of cross-sectionally demeaned, N-period discounted future stock return on cross-sectionally demeaned BE/ME. The second column shows the simple predictive regression coefficient of cross-sectionally demeaned, N-period discounted future profitability (ROE) on cross-sectionally demeaned BE/ME. The third column shows the simple predictive regression coefficient of cross-sectionally demeaned, N-period- in-the-future discounted BE/ME on cross-sectionally demeaned BE/ME.

in $SULO &RUUHVSRQGHQ�H &KULVWRSKHU 3RON.HOORJJ*UDGXDWH 6�KRRO RI 0DQDJHPHQW 1RUWKZHVWHUQ 8QLYHUVLW\

"... In PAGE 17: ... Similarly, a variance decomposition conditional on the BE/ME ratio may highlight asymmetries in the amount of transitory variation due to expected returns, perhaps as a result of short sale constraints that prevent liquidity providers from selling stocks with very low expected returns. Therefore, we sort firms into three size groups based on NYSE breakpoints to examine how the variance decomposition varies with market capitalization ( Table4 Panel A) and we sort firms into three groups based on firm BE/ME (Table 4 Panel B) to examine how the decomposition depends on BE/ME. The decomposition conditional on size indicates that the book-to-market ratios have less information concerning differences in expected returns among large firms than small or medium-sized firms.... In PAGE 17: ... Similarly, a variance decomposition conditional on the BE/ME ratio may highlight asymmetries in the amount of transitory variation due to expected returns, perhaps as a result of short sale constraints that prevent liquidity providers from selling stocks with very low expected returns. Therefore, we sort firms into three size groups based on NYSE breakpoints to examine how the variance decomposition varies with market capitalization (Table 4 Panel A) and we sort firms into three groups based on firm BE/ME ( Table4 Panel B) to examine how the decomposition depends on BE/ME. The decomposition conditional on size indicates that the book-to-market ratios have less information concerning differences in expected returns among large firms than small or medium-sized firms.... In PAGE 18: ...39 for the largest third of the sample. Panel B of Table4 allows the decomposition to vary with BE/ME. Largely by construction, the medium-BE/ME third has an average cross-sectional variance of log BE/ME of just 0.... In PAGE 22: ...When we add both the median BE/ME ratio and the default spread to the return-predictive regressions in Table4 , we find that the value spread remains significant (coefficient of .394, t-statistic of 3.... ..."

### Table 6: No Short-sales, 4-year Estimation Period, Portfolio Properties

"... In PAGE 16: ... Moreover, the performance in the last couple of years has been notably negative.19 Table6 sets out growth and volatility statistics for the classic and ridge short-sales not permitted portfolios and for each of the benchmarks, for the 4-year estimation periods. It is clear from Table 6 that the growth oriented strategies, while having considerably higher growth rates than the benchmark strategies, also have much higher volatility than the benchmark MVP and 15% growth and equally weighted strategies.... In PAGE 16: ...19 Table 6 sets out growth and volatility statistics for the classic and ridge short-sales not permitted portfolios and for each of the benchmarks, for the 4-year estimation periods. It is clear from Table6 that the growth oriented strategies, while having considerably higher growth rates than the benchmark strategies, also have much higher volatility than the benchmark MVP and 15% growth and equally weighted strategies. The direct relationship between growth and volatility is also evident in the 3-year and 5- year estimation period as Table 5 shows.... In PAGE 16: ... At the other end of the spectrum is the maximally diversified, equally weighted portfolio with unit length. Table6 shows that the classic no short-sales allowed, growth optimal portfolio with an average length of 4.36, is at the lower end of the diversity spectrum.... In PAGE 17: ... Page - 16 - Table6 shows that this portfolio contains on average only 1.71 assets in each period.... In PAGE 18: ... One can see that this is the case to a degree with the no short-sales allowed growth portfolios. Table6 shows that the maintenance of this growth optimal portfolio strategy would have required on average a turnover of stock of about 12% per month. The transaction costs associated with any strategy that turns over 12% of a portfolio per month are considerable, and will significantly lower the effective rate of portfolio growth.... In PAGE 18: ...35 to facilitate comparison with the short-sales allowed results. Predictably, Table6 reveals that the no short-sales allowed, ridge, growth portfolio performance lies somewhere between the performance of the classic growth optimal portfolio and the equally weighted portfolio. The no short-sales allowed, ridge, 22 For example, the cost of portfolio adjustment would be 2.... ..."

### Table 4. 4.4. Scenario III: no transaction cost but short sale is permitted

in Arbitrage pricing theory-based Gaussian temporal factor analysis for adaptive portfolio management

### Table III Constraint probabilties p2|1, auto sales data seta Probability Numerical Timeb

### Table III Constraint probabilties p2|1, auto sales data seta Probability Numerical Timeb

### Table 9: Sales and Wage Elasticities of the Demand for Liquid Assets (OLS) independent dependent variable: log (cash amp; short-term investments) variables

1997

"... In PAGE 39: ... The additions include holdings of government bonds, time deposits, and commercial paper. Table9 displays OLS estimates of the money demand equation (3) apos;, using the log of quot;cash and short-term investments quot; as the dependent variable. Only year effects and log sales are included as regressors in column 1; the sales elasticity for liquid assets (0.... In PAGE 40: ...i. for the years 1969-90 (5) year effects are estimated in every regression (6) N is the number of firm-year cells included in the regression (7) standard errors are displayed in parentheses It appears from Table9 that the wage and sales elasticities of the demand for the more broad... ..."