# Multiple alignments

If I have a string of equations whose equal signs are aligned and one of the equations is too long and needs to span multiple lines, how do I create a second alignment. For instance, in the following example

\begin{align*}
f(x)&=eqn1\\
&=eqn2_part1+eqn2_part2\\
&           +eqn2_part3\\
&=eqn3\\
&=eqn4\\
\end{align*}


How do I align the equals signs and the plus signs.

-

Using boxes via \phantom is also an option, if alignment characters like && get in the way for long equations:

\documentclass{article}
\usepackage{amsmath}% http://ctan.org/pkg/amsmath
\begin{document}
\begin{align*}
f(x)&=\text{eqn}_1\\
&=\text{eqn}_2\text{-part}_1+\text{eqn}_2\text{-part}_2\\
&\phantom{{}=\text{eqn}_2\text{-part}_1}{}+\text{eqn}_2\text{-part}_3\\
&=\text{eqn}_3\\
&=\text{eqn}_4\\
\end{align*}
\end{document}


A correction for the spacing around binary operators ({}+) and relations ({}=) is required though. See Herbert's mathmode document for more on AMS environments and alignment.

-

You could use alignat which is also from the amsmath package.

\documentclass{article}

\usepackage{amsmath}

\begin{document}

\begin{alignat*}{2}
f(x)&=eqn1                    &&   \\
&=eqn2_part1              && +eqn2_part2 \\
&                         && +eqn2_part3\\
&=eqn3                    &&   \\
&=eqn4                    &&   \\
\end{alignat*}

\end{document}


You'll notice that the alignat environment takes an argument- which is (quoting from the documentation)

the number of “equation columns”: count the maximum number of &s in any row, add 1 and divide by 2.

-
When I compile I get a large gap between eqn2_part1 and the plus signs when the length of eqn3 and 4 are longer than eqn2_part1. –  richtera Jun 1 '12 at 16:18
@richtera perhaps you could add your code to your question so that folks can play with it? –  cmhughes Jun 1 '12 at 16:31

Please try breqn package. Then use \begin{dmath}...\end{dmath} environment. For example

\documentclass[12pt]{article} %
\usepackage{breqn}
\begin{document}
\begin{dmath}
g\left[Pr(unrest_{im}=1)\right]=\alpha_{0i}+\alpha_{1}.\left|price\_shock_{im}\right|+\beta.{year}+\gamma.region.\left|price\_shock_{im}\right|+\delta.{region}+\epsilon_{im}

\end{dmath}
\end{document}


-