# Text wrapping of text in multirow

I want to wrap rotated text which goes out of the lines. The line I want to fix is given below:

\parbox[t]{2mm}{\multirow{3}{*}{\rotatebox[origin=c]{90}{Summary of the monthly returns for the portfolios}}}


\documentclass[11pt]{article}
\usepackage{setspace,amssymb,amsmath}
\usepackage[tablename=TABLE,labelsep=newline,
aboveskip=0pt,font=bf,justification=centering]{caption}
\usepackage{booktabs,tabularx}
\usepackage[margin=1in]{geometry}
\usepackage[autostyle]{csquotes}
\usepackage[table]{xcolor}
\usepackage{multirow}
\usepackage{array}
\usepackage{graphicx}

\newcolumntype{Y}{>{\centering\arraybackslash}X}
\newcommand{\gmc}[3]{\multicolumn{#1}{@{}#2@{}}{#3}}

\doublespacing

\begin{document}

\newpage
\begingroup % keep any font size changes local to group
%\setlength\tabcolsep{2pt} % default value: 6pt
\captionof{table}{Portfolio-Level Investment Analysis: Future Risk-Adjusted REIT Returns for Portfolios Constructed by Quarterly Sort on Regional Accounting Profitability Changes}
\singlespacing
\footnotesize
\setlength\tabcolsep{6pt} %default
\renewcommand{\arraystretch}{1.2}

\begin{tabularx}{\linewidth}{@{}*{8}{Y}@{}}
\toprule
Portfolio:  &   & Portfolio Alpha&  MKTRF&  SMB &HML&   UMD & Adj. $R^2$ \\\midrule
1 (Lowest) &    Coefficient&    -0.0004 &0.770& 0.510&  0.811&  -0.175& 0.659\\
&t-statistic&   -1.800& 11.457& 5.945&  9.471&  -3.395  \\
2&  Coefficient&    0.0036& 0.759&  0.466&  0.709&  -0.054& 0.626\\
&t-statistic&   1.343&  11.522& 5.102&  8.419&  -1.063  \\
3&  Coefficient&    0.0041& 0.676&  0.671&  0.809&  -0.265& 0.604\\
&t-statistic&   1.302&  8.588&  6.146&  8.030&  -4.367  \\
4&  Coefficient&    0.0031& 0.760&  0.530&  0.872&  -0.183& 0.595\\
&t-statistic&   0.961&  9.681&  5.281&  8.720&  -3.047  \\
5 (Highest)&    Coefficient&    0.0068& 0.614&  0.642&  0.778&  -0.220& 0.571\\
&t-statistic&   2.810&  7.957&  5.999&  7.876&  -3.694   \\\hline
\parbox[t]{2mm}{\multirow{3}{*}{\rotatebox[origin=c]{90}{Summary of the monthly returns for the portfolios}}} & High minus low return&  0.0072   \\
&t-statistic&   \cellcolor{gray!25}\textbf{6.036}   \\\cmidrule(l){2-3}
&Hedge return for a zero-cost investment portfolio& 0.0072  \\
&t-statistic&   \cellcolor{gray!25}\textbf{3.080}   \\\cmidrule(l){2-3}
&Trend& 0.0014  \\
&t-statistic&   2.800   \\
&Adjusted $R^2$ &   0.631\\
\end{tabularx}

\footnotesize \hrule \medskip
Notes: The table reports results from the portfolio-level investment strategy of future REIT stock returns for portfolios constructed by quarterly sorting on regional accounting profitability changes. Abnormal returns are the intercepts (i.e., alphas) from monthly time-series regressions of future excess returns for the portfolios conditioned on past geographic accounting information. In particular, I estimate a time-series model at the portfolio level to obtain portfolio intercepts, as described in Equation (6). To test for abnormal hedge returns, I construct a zero-cost investment portfolio that longs the portfolio with the sharpest regional accounting profitability increase (highest portfolio) and shorts the portfolio with the sharpest regional accounting profitability decrease (lowest portfolio). I then regress the zero-cost portfolio’s returns on the same-period factor returns. Raw stock returns are from the CRSP Monthly Stock File, adjusted for delisting returns. The risk-free rate and Fama-French-Carhart factors are from the Fama-French Portfolios and Factors dataset available on WRDS. Accounting data are from Compustat North America Fundamentals Quarterly. To ensure the feasibility of investing in REITs when housing data exist across all regions, this analysis employs quarterly observations for which there are real estate index data for all regions. Accordingly, this analysis uses quarterly data beginning in Q4:1999, because the first month for which real estate regional data are available for all regions is January 2000 (the last region for which data are available is Dallas, TX). The sample consists of U.S. stocks in the intersection of CRSP and Compustat with data available to calculate quarterly changes to regional accounting profitability, with quarterly profitability changes beginning Q4:1999, and with future stock returns available as of March 2015. Because the analyses require future returns, the investment construction quarters are Q4:1999-Q2:2014, with monthly stock returns beginning in March 2000 (one month after all regional profitability changes are available) and ending in November 2014 (last of the three months of stock returns-for September, October, and November-for investing based on the Q2:2014 regional accounting indices available in middle of August 2014).
\medskip
\hrule
\endgroup

\end{document}


In my experience, readers truly dislike having to crane their necks in order to read rotated material -- especially if it's just a line or two that's rotated. For the table at hand, it doesn't appear to be necessary to rotate the string "Summary of the monthly returns for the portfolios". Typesetting it as a normal, i.e., non-rotated header seems just fine.

I would like to recommend that you typeset the numbers in columns 3 thru 8 so that (a) they are aligned on their respective decimal markers and (b) negative numbers are prefixed with a typographic minus symbol rather than a simple "dash". One way to achieve this is to load the siunitx package and to use that package's S column type for the columns in question. Separately, I would also strive not to over-do emphasis: Use bold or shading to attract attention to the crucial cells -- but don't employ both methods simultaneously.

\documentclass[11pt]{article}
\usepackage{setspace,amssymb,amsmath}
\usepackage[tablename=TABLE,labelsep=newline,
aboveskip=0pt,font=bf,justification=Centering,
skip=0.5\baselineskip]{caption}
\usepackage{booktabs,tabularx}
\usepackage[margin=1in]{geometry}
\usepackage[autostyle]{csquotes}
\usepackage[table]{xcolor}
\usepackage{array,graphicx,siunitx,ragged2e}

\newcolumntype{Y}{>{\centering\arraybackslash}X}
\newcommand{\gmc}[3]{\multicolumn{#1}{@{}#2@{}}{#3}}

\doublespacing

\begin{document}

\begin{table}[p!]
\singlespacing
\footnotesize
\renewcommand{\arraystretch}{1.2}

\caption{Portfolio-Level Investment Analysis: Future Risk-Adjusted REIT
Returns for Portfolios Constructed by Quarterly Sort on Regional
Accounting Profitability Changes}

\begin{tabularx}{\linewidth}{@{} ll S[table-format=-1.4]
S[table-format=2.3] YY S[table-format=-1.3] Y  @{}}
\toprule
Portfolio  &
& \multicolumn{1}{Y}{Portfolio Alpha}
& \multicolumn{1}{Y}{MKTRF}&  SMB &HML
& \multicolumn{1}{Y}{UMD} & Adj.\ $R^2$ \\
\midrule
1 (Lowest) &    Coefficient&    -0.0004 &0.770& 0.510&  0.811&  -0.175& 0.659\\
&t-statistic&   -1.800& 11.457& 5.945&  9.471&  -3.395  \\
2&  Coefficient&    0.0036& 0.759&  0.466&  0.709&  -0.054& 0.626\\
&t-statistic&   1.343&  11.522& 5.102&  8.419&  -1.063  \\
3&  Coefficient&    0.0041& 0.676&  0.671&  0.809&  -0.265& 0.604\\
&t-statistic&   1.302&  8.588&  6.146&  8.030&  -4.367  \\
4&  Coefficient&    0.0031& 0.760&  0.530&  0.872&  -0.183& 0.595\\
&t-statistic&   0.961&  9.681&  5.281&  8.720&  -3.047  \\
5 (Highest)&    Coefficient&    0.0068& 0.614&  0.642&  0.778&  -0.220& 0.571\\
&t-statistic&   2.810&  7.957&  5.999&  7.876&  -3.694   \\
\midrule
\multicolumn{8}{@{}l@{}}{Summary of monthly portfolio returns} \\[1ex]
\multicolumn{8}{@{}l@{}}{High minus low return}\\
&Coefficient & 0.0072   \\
&t-statistic & \cellcolor{gray!25} 6.036    \\
\multicolumn{8}{@{}l@{}}{Hedge return for a zero-cost investment portfolio}  \\
&Coefficient & 0.0072  \\
&t-statistic & \cellcolor{gray!25} 3.080    \\
Trend
&Coefficient & 0.0014  \\
&t-statistic & 2.800   \\[1ex]
&Adj.\ $R^2$ & 0.631   \\
\bottomrule
\end{tabularx}

\medskip
Notes: The table reports results from the portfolio-level investment strategy of future REIT stock returns for portfolios constructed by quarterly sorting on regional accounting profitability changes. Abnormal returns are the intercepts (i.e., alphas) from monthly time-series regressions of future excess returns for the portfolios conditioned on past geographic accounting information. In particular, I estimate a time-series model at the portfolio level to obtain portfolio intercepts, as described in Equation (6). To test for abnormal hedge returns, I construct a zero-cost investment portfolio that longs the portfolio with the sharpest regional accounting profitability increase (highest portfolio) and shorts the portfolio with the sharpest regional accounting profitability decrease (lowest portfolio). I then regress the zero-cost portfolio's returns on the same-period factor returns. Raw stock returns are from the CRSP Monthly Stock File, adjusted for delisting returns. The risk-free rate and Fama-French-Carhart factors are from the Fama-French Portfolios and Factors dataset available on WRDS\@. Accounting data are from Compustat North America Fundamentals Quarterly. To ensure the feasibility of investing in REITs when housing data exist across all regions, this analysis employs quarterly observations for which there are real estate index data for all regions. Accordingly, this analysis uses quarterly data beginning in Q4:1999, because the first month for which real estate regional data are available for all regions is January 2000 (the last region for which data are available is Dallas,~TX)\@. The sample consists of U.S. stocks in the intersection of CRSP and Compustat with data available to calculate quarterly changes to regional accounting profitability, with quarterly profitability changes beginning Q4:1999, and with future stock returns available as of March 2015. Because the analyses require future returns, the investment construction quarters are Q4:1999--Q2:2014, with monthly stock returns beginning in March 2000 (one month after all regional profitability changes are available) and ending in November 2014 (last of the three months of stock returns---for September, October, and November---for investing based on the Q2:2014 regional accounting indices available in middle of August 2014).
\end{table}

\end{document}

• I added a higher quality picture for you. – Henri Menke Jun 20 '16 at 9:22