Context :

Aims : The radio-quiet quasar PG 1416 - 129 ( z = 0.129 ) exhibits atypical optical and X-ray properties .
Between 1990 and 2000 , in response to its optical continuum decrease , the “ classical ” broad component of H \beta almost completely disappeared , with a factor of 10 decrease in the line flux .
In addition , the width of the broad component of the H \beta line decreased significantly from 4000 km s ^ { -1 } to 1450 km s ^ { -1 } .
In the X-ray band , this object was observed by Ginga in 1988 to have the hardest quasar photon index , with \Gamma =1.1 \pm 0.1 .
We present an XMM-Newton /EPIC observation of PG 1416 - 129 performed in July 2004 .

Methods : We analyze the time-averaged pn spectrum of this quasar , as well as perform time-resolved spectroscopy .

Results : We find that during the present XMM-Newton observation , PG 1416 - 129 still has a rather hard photon index , both in the soft ( 0.2–2 keV ) and hard ( 2–12 keV ) energy ranges , compared to radio-quiet quasars ( BLS1 and NLS1 ) but compatible with the photon index value found for radio-loud quasars .
This object also shows long-term luminosity variability over 16 years by a factor of three with a variation of photon index from \sim 1.2 to \sim 1.8 .
In the soft energy band ( 0.2–2 keV ) , we found a very weak soft X-ray excess compared to other RQ quasars .
The whole time averaged spectrum is fit very well either by X-ray ionized reflection from the accretion disk surface , by a warm absorber-emitter plus power-law , or by a smeared absorption/emission from a relativistic outflow .
While no constant narrow Fe K line at 6.4 keV is observed , we find the possible presence of two non-simultaneous transient iron lines : a redshifted narrow iron line at about 5.5 keV ( 96.4 \% confidence level according to multi-trial Monte-Carlo simulations ) at the beginning of this observation and the appearance of a line at 6.3–6.4 keV ( 99.1 \% c.l . )
at the end of the observation .
These variable lines could be generated by discrete hot-spots on the accretion disk surface .

Conclusions :