Error in PostgreSQL function with dynamic query for update

Adrian already pointed out the immediate syntax error.
But there is more. Consider this rewrite:

CREATE OR REPLACE FUNCTION mod_medicamento(_tabla char, _nombre text, _operacion char, _porcentaje float) -- ①
  RETURNS void
  LANGUAGE plpgsql AS
$func$
DECLARE
/*
$1 table (L=laboratorio, P=proovedor(supplier), M=medicamento
$2 name of lab, supplier, or product
$3 paramater for type of operation (A for increment price, D for decrement price)
$4 for percentage(0.01 to 0.99)
*/
   _nombre_tabla text;
   _nombre_campo text;
   _calculo text;
BEGIN
   -- enforce: $4 for percentage(0.01 to 0.99)  -- ②
   IF _porcentaje > 0 AND _porcentaje < 1 THEN  -- actually in range (0, 1)
      -- all good
   ELSE -- also catches null
      RAISE EXCEPTION '_porcentaje must be in range [0, 1). Was: %', _porcentaje;
   END IF;
   
   CASE _tabla  -- ③
   WHEN 'L' THEN
      _nombre_tabla := 'laboratorio';
      _nombre_campo := 'laboratorio';
   WHEN 'P' THEN
      _nombre_tabla := 'proovedor';
      _nombre_campo := 'proovedor';
   WHEN 'M' THEN
      _nombre_tabla := 'medicamento';
      _nombre_campo := 'nombre';
   ELSE
     RAISE EXCEPTION 'selecciona una opcion valida (L/P/M)';
   END CASE;

   CASE _operacion
   WHEN 'A' THEN
      _calculo := '+';
   WHEN 'D' THEN
      _calculo := '-';
   ELSE
     RAISE EXCEPTION 'selecciona una opcion valida de operacion(A/D)';
   END CASE;

   EXECUTE format(
-- RAISE NOTICE '%'  , format(  -- use instead of EXECUTE for debugging!
$q$
UPDATE medicamento m
SET    precio = m.precio * (1 %1$s %2$s)  -- ④, ⑤
FROM   %3$I t                             -- ⑥
WHERE  t.id_%4$s LIKE %5$L
AND    m.id_%3$s = t.id_%3$s
$q$
    , _calculo, _porcentaje, _nombre_tabla, _nombre_campo, _nombre);
END
$func$;

Generates and executes a query of the form:

UPDATE medicamento m
SET    precio = m.precio * (1 + 0.123)
FROM   proovedor t
WHERE  t.id_proovedor LIKE 'nombre'
AND    m.id_proovedor = t.id_proovedor

① Use of ALIAS is discouraged. Use named parameters instead.
I like to prefix parameter and variable names with an underscore (_) to avoid naming conflicts with column names, but that's optional.

② Actually enforce the allowed range for _porcentaje. (Also disallowing 0 which wouldn't change anything.)

③ A "switched" CASE is more efficient & elegant for the use case. Example:

• Display column name with max value between several columns

④ Simplified computation. precio + precio * 0.123 --> precio * (1 + 0.123)

⑤ Your original concatenation did not work at all since the format specifier %L in format() adds single quotes that break the syntax.

⑥ Use a FROM clause in the UPDATE instead of a correlated subquery. Your original query raises an exception if more than one row matches. Also: faster.

Since you are not returning anything, you may want a PROCEDURE instead of the FUNCTION. But either works. See:

• When to use stored procedure / user-defined function?

Dynamic SQL like this is potentially open to SQL injection. But while you don't concatenate user input directly, you are safe.